The integration of dynamic, sparse structures like Mixture-of-Experts (MoE) with parameter-efficient adapters (e.g., LoRA) is a powerful technique for enhancing Large Language Models (LLMs). However, this architectural enhancement comes at a steep cost: despite minimal increases in computational load, the inference latency often skyrockets, leading to decoding speeds slowing by over 2.5 times. Through a fine-grained performance analysis, we pinpoint the primary bottleneck not in the computation itself, but in the severe overhead from fragmented, sequential CUDA kernel launches required for conventional dynamic routing. To address this challenge, we introduce AdaFuse, a framework built on a tight co-design between the algorithm and the underlying hardware system to enable efficient dynamic adapter execution. Departing from conventional layer-wise or block-wise routing, AdaFuse employs a token-level pre-gating strategy, which makes a single, global routing decision for all adapter layers before a token is processed. This ``decide-once, apply-everywhere'' approach effectively staticizes the execution path for each token, creating an opportunity for holistic optimization. We capitalize on this by developing a custom CUDA kernel that performs a fused switching operation, merging the parameters of all selected LoRA adapters into the backbone model in a single, efficient pass. Experimental results on popular open-source LLMs show that AdaFuse achieves accuracy on par with state-of-the-art dynamic adapters while drastically cutting decoding latency by a factor of over 2.4x, thereby bridging the gap between model capability and inference efficiency.

Dataset distillation has achieved remarkable progress as an effective approach for data compression. However, real-world data often comes from diverse domains, leading to potential mismatches between the domains of synthesized images and those of the evaluation set. Existing methods primarily assume domain alignment between them, which limits their generalization ability in the above cross-domain scenarios. In this paper, we aim to ensure that images synthesized from known domains maintain robust performance on unseen domains and propose a novel framework called Channel-masked Asymmetric Distribution Matching (CADM). During asymmetric distribution matching, domain-sensitive channels of real data are selectively masked at different layers to extract domain-invariant features that guide synthetic data optimization. To further improve synthetic data representation, we introduce a class-focused domain-agnostic regularization to capture class-relevant knowledge while ignoring domain-specific information. Experiments show that our method produces domain-robust synthetic data and substantially improves generalization performance on unseen domains.

Aggregated time series are widely used in business and economics, where top-level sequences (e.g., category sales) aggregated from underlying sequences (e.g., individual items) often exhibit clearer trends and are therefore typically the primary focus of forecasting tasks. However, treating top-level sequences as ordinary multivariate time series is inappropriate in the presence of coupled aggregation constraints. The core challenge arises in coupled aggregation structures, where a single underlying sequence contributes to multiple top-level sequences, as simple nonnegativity constraints of underlying sequences induce highly complex constraints among top-level sequences. Existing methods fail to achieve high accuracy while satisfying these constraints. To address this, we propose ProCAST, a projection-based framework that adjusts forecasts from any multivariate base model to satisfy coupled aggregation constraints. By introducing virtual underlying sequences and leveraging orthogonal and oblique projection, our method ensures that the top-level forecasts are feasible without explicitly deriving complex constraints. Theoretically, we prove that the proposed method guarantees improved accuracy under distance-based loss functions. Experiments on real-world datasets show that our method completely eliminates constraint violations while achieving higher accuracy than current state-of-the-art approaches.

LLM agents increasingly rely on tool calling, and the Model Context Protocol (MCP) standardizes it between agents and tool providers, reducing integration cost and driving rapid growth in tool scale. Yet a standardized interface does not make tool access work at production scale: legacy services are not MCP-callable, fast protocol evolution creates compatibility cost, large tool sets exhaust the context window, and stateful sessions complicate load balancing. We solve these with a shared control point, a centralized MCP Gateway System that makes MCP operational at cloud scale. The gateway breaks the direct-connect data plane and consolidates legacy API integration, protocol bridging, access control, and session-aware routing, while scaling out elastically at low per-call overhead. It scales agent tool access to thousands of cloud operations.

Large language model (LLM) training is prone to anomalies due to its long duration and large scale, which can lead to significant performance degradation or even training crashes. Due to the synchronization nature of LLM training, anomalies exhibit the cascading effect, making their diagnosis challenging. Existing approaches rely on collecting communication operator information via code instrumentation, which yields only coarse-grained monitoring data and requires modifications to training code or communication libraries. We propose Pulse, a fine-grained, non-intrusive, and easy-to-deploy monitoring system. Our key idea is to enable fine-grained monitoring via traffic measurement. Pulse conducts microsecond-level RDMA traffic measurement on NICs, and transforms flow-level measurements into communication operator measurements, thereby enabling fine-grained and non-intrusive monitoring. We deploy Pulse on a testbed with 64 H200 GPUs and evaluate its anomaly localization capability under common failure scenarios. Pulse achieves machine-level localization in 10 out of 12 scenarios, while existing methods succeed in only 4 and even misdiagnose 2 of the remaining scenarios. Additionally, Pulse achieves over 90% precision and 100% recall, supports up to 2000 concurrent RDMA flow measurements per NIC, and imposes negligible overhead on training performance, making it a practical solution for real-world LLM training environments.

Real-time LLM interactions demand streamed token generations, where text tokens are progressively generated and delivered to users while balancing two objectives: responsiveness (i.e., low time-to-first-token) and steady generation (i.e., required time-between-tokens). Standard LLM serving systems suffer from the inflexibility caused by non-preemptive request scheduling and reactive memory management, leading to poor resource utilization and low request processing parallelism under request bursts. Therefore, we present TokenFlow, a novel LLM serving system with enhanced text streaming performance via preemptive request scheduling and proactive key-value (KV) cache management. TokenFlow dynamically prioritizes requests based on real-time token buffer occupancy and token consumption rate, while actively transferring KV cache between GPU and CPU memory in the background and overlapping I/O with computation to minimize request preemption overhead. Extensive experiments on Llama3-8B and Qwen2.5-32B across multiple GPUs (RTX 4090, A6000, H200) demonstrate that TokenFlow achieves up to 82.5% higher effective throughput (accounting for actual user consumption) while reducing P99 TTFT by up to 80.2%, without degrading overall token throughput.

The explosive growth of modern short-video platforms like WeChat Channels — approaching 1.4 billion monthly active users (MAU) while serving over one billion queries per second (QPS) — has exposed fundamental limitations in conventional content delivery network (CDN) architectures, particularly their inability to handle highly dynamic content popularity patterns. The traditional cache-centric approach proves to be catastrophic when faced with ephemeral content exhibiting a "million-item waist" distribution. This phenomenon results in severe cache pollution (manifesting as 5–40% origin miss ratios) and introduces prohibitive back-to-origin bandwidth costs. Our TritonFlow overcomes these challenges by introducing access point scheduling as a novel origin traffic compensator, which reduces the operational cost of back-to-origin fetches by enhancing CDN cache affinity. Through analysis of production traces (peak >1B QPS), we demonstrate that TritonFlow reduces scheduling fluctuations by 96% and decreases back-to-origin traffic caused by non-compulsory cache misses by 35%, all while maintaining sub-200ms fetch latency.

Range-top-k queries retrieve the top-k elements within an arbitrary subrange of a large array and are a key primitive in real-time analytics. Unlike one-shot top-k selection, practical deployments issue large volumes of queries over varying and often overlapping ranges, frequently interleaved with streaming updates. In this setting, applying conventional GPU top-k kernels per query is inefficient: each query triggers range rescans or O(n)-scale passes that overwhelm HBM bandwidth, thrash on-chip caches, and provide little reuse across overlapping windows.

This paper introduces Vistar, a novel self-supervised framework for inertial measurement unit (IMU) signal perception designed for large language models (LLMs). Unlike visual data, IMU signals are high-frequency time series with low interpretability, making manual annotation with natural language particularly challenging. Even when using vision-language models (VLMs) to describe events in videos synchronized with IMU signals, a semantic gap remains between high-level visual semantics and low-level IMU vibrations. The core idea of Vistar is to achieve accurate IMU signal perception through collaborations between offline cross-modal alignment and online retrieval-augmented generation. During offline training, Vistar uses pretrained vision and language encoders as anchors to learn IMU encoders via hierarchical cross-modal contrastive learning, establishing both inter- and intra-sample alignment. Given that the enhanced training strategy still fails to achieve precise alignment between IMU and text, during online inference, Vistar further employs a retrieval-augmented generation mechanism to generate distilled textual descriptions from similar text filtered based on structural relations of their paired IMU samples. Extensive evaluations on three multimodal datasets demonstrate that Vistar consistently outperforms state-of-the-art (SOTA) baselines by up to 57.45% in IMU-to-text retrieval and improves the generated text similarity with ground truths in IMU perception by up to 31.90%.

RFID has revolutionized item-level intelligence in IoT ecosystems, yet static localization with passive tags remains challenging due to multipath interference inherent in RF signals. We present LuxTag, the first system to enable visible light-based sensing and localization using standard, commercial RFID tags by transforming them into ambient light sensors. Our key insight leverages the discovery that photon-induced leakage currents in passive RFID ICs modulate their persistence time (i.e., the duration a tag remains operational after RF excitation ceases) proportional to ambient illuminance. LuxTag introduces two innovations: (i) a first-principles model characterizing how ambient light alters tag persistence time, enabling battery-free light sensing without hardware modifications; (ii) a differential measurement technique and zero-shot calibration method to isolate light effects and autonomously derive tag parameters, ensuring robust and accurate static localization system using COTS RFID infrastructure. Extensive experiments demonstrate that LuxTag achieves a mean light intensity error of 3.6 lux and 60.7% improvement over state-of-the-art static RFID localization. By synergizing the ubiquity of RFID with the multipath resilience of optical sensing, LuxTag opens new avenues for static RFID localization in smart warehouses, retails, and beyond.

Machine learning models are widely integrated into modern mobile apps to analyze user behaviors and deliver personalized services. Ensuring low-latency on-device model execution is critical for maintaining high-quality user experiences. While prior research has primarily focused on accelerating model inference with given input features, we identify an overlooked bottleneck in real-world on-device model execution pipelines: extracting input features from raw application logs. In this work, we explore a new direction of feature extraction optimization by analyzing and eliminating redundant extraction operations across different model features and consecutive model inferences. We then introduce AutoFeature, an automated feature extraction engine designed to accelerate on-device feature extraction process without compromising model inference accuracy. AutoFeature comprises three core designs: (1) graph abstraction to formulate the extraction workflows of different input features as one directed acyclic graph, (2) graph optimization to identify and fuse redundant operation nodes across different features within the graph; (3) efficient caching to minimize operations on overlapping raw data between consecutive model inferences. We implement a system prototype of AutoFeature and integrate it into five industrial mobile services spanning search, video and e-commerce domains. Online evaluations show that AutoFeature reduces end-to-end on-device model execution latency by 1.33 × -3.93 × during daytime and 1.43 × -4.53 × at night.

The development of large language models (LLMs) has catalyzed unprecedented demand on the computing network, specifically for large-scale, few-hops, and low-latency, which directly underpin LLM task efficiency. However, mainstream topologies such as Clos suffer from costs and latency, while topologies with good scalability have symmetric or collective communication issues. In order to achieve a favorable balance among design metrics, we propose a novel topology named the Balanced Sparse Tree (BST), which is a topology characterized by symmetric design and sparse connections, motivated by hypergraph theory and Steiner Systems. Its degree-diameter upper-bound approaches the Moore Bound for Bipartite Biregular graphs, larger than other known dia-meter-2 topologies. Furthermore, we incorporate differentiated routing, deadlock freedom, and topology-affined deployment into BST. Testbed experiments, simulations, together with modeling analysis, demonstrate the superiority of BST over the state-of-the-art in network scale, latency, bandwidth, and cost. With equivalent scales, BST outperforms Clos with a 50% cost reduction while maintaining comparable performance for AI workloads. Furthermore, BST delivers a 3.9%–11.8% gain in collective communications and has 13.4% improvement over state-of-the-art topologies.

The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA's NVL72/576 and Huawei's CloudMatrix384, introduces critical challenges beyond raw interconnect bandwidth. While these systems provide unified global address spaces and high-bandwidth fabrics, their full potential for sparse MoE communication is hindered by three fundamental bottlenecks: (1) Strict execution serialization imposed by coarse-grained Bulk Synchronous Parallel (BSP) orchestration of interdependent communication phases; (2) Prohibitive synchronization overhead that fails to scale alongside high interconnect bandwidth; and (3) Severe load imbalance resulting from distance-agnostic scheduling of irregular token traffic. To eliminate these bottlenecks, we introduce UBEP (Unified-Bus Expert Parallelism), a production-ready communication library that rethinks MoE's All-to-All primitives for modern superpod architectures. Through large-scale experiments, UBEP reduces All-to-All latency by up to 52.4% and MoE inference Time Per Output Token (TPOT) by up to 11.1%.

Low Earth Orbit (LEO) Synthetic Aperture Radar (SAR) satellites conventionally rely on dedicated communication links, which impose prohibitive hardware, spectrum, and power overhead as satellite constellations scale. This paper presents CommSAR, a novel system that reuses existing SAR imaging waveforms to enable bidirectional communication without modifying satellite hardware or compromising imaging performance. For the downlink, data are embedded by modulating the starting frequency offset of the imaging waveform, preserving the waveform structure and imaging quality. For the uplink, we propose a compact, low-cost programmable metasurface to replace conventional large, expensive antennas, significantly lowering the barrier for dense ground station deployment. To handle extreme satellite dynamics, we employ an opposite-slope waveform as a pilot to compensate for mobility-induced effects. We implement a ground station prototype of CommSAR and validate its performance using an in-orbit commercial SAR satellite and a UAV SAR platform. Experimental results show that CommSAR preserves imaging performance without degradation while achieving downlink and uplink data rates of up to 105 kbps and 112 kbps, respectively, significantly outperforming the state of the art and demonstrating utility-grade performance.

Embedded AI applications usually require compact on-device models that can continually adapt to new tasks. However, recent studies have revealed that neural networks trained on non-stationary data streams gradually lose their ability to adapt to new tasks, a phenomenon known as plasticity loss. Moreover, to enable neural networks to run on resource-constrained embedded devices, model pruning is commonly applied for model compression, which may further affect their plasticity. To conduct efficient model adaptation on new tasks on embedded devices, we propose Plasticity-aware Continual Pruning (PaCP), a novel framework that operates in two stages. First, a pre-deployment stage uses a plasticity-aware strategy to prune the model while optimizing its initial structure for future adaptability. Second, during continual learning, the model's capacity is temporarily expanded at task boundaries to efficiently learn new information, before plasticity-aware pruning restores its compact form. Extensive experiments on multiple continual learning benchmarks demonstrate that PaCP significantly outperforms existing plasticity-maintenance methods and, remarkably, even surpasses non-pruned models lacking explicit plasticity preservation.

Collaborative learning between edge servers (e.g., base stations) and end devices (e.g., drones) enables simultaneous model training in web applications through knowledge sharing. The resulting models effectively reduce service latency. However, existing approaches either assume isomorphic models on edge servers and end devices or incur substantial transmission overhead when training. Moreover, edge servers are often unable to access data from end devices on time due to long-distance constraints or strict data privacy regulations. This paper proposes a Prototype Augmentation-based Edge-end Collaborative Learning method (PAECL). It simultaneously trains heterogeneous edge and end models in the absence of data on edge servers by transmitting only augmented class-wise feature vectors (prototypes), significantly reducing communication overhead compared to sharing models, data, or logits. Specifically, on end devices, prototype-implied latent knowledge is augmented via local prototype contrast and global prototype alignment. On edge servers, prototypes are further augmented to produce bounded virtual vectors by mixing them with random noise, and the augmented prototypes are then delivered to generative models to provide data during edge model training. Through simulations and field experiments, PAECL achieves the highest accuracy for edge and end models under limited training resources and reduces the transmission burden by at least 297 times compared to existing edge-end heterogeneous learning methods.

The security of DNS over HTTPS (DoH) faces escalating challenges driven by advances in quantum computing, adversarial artificial intelligence, and increasingly sophisticated attack strategies targeting encrypted communication channels. While existing protection mechanisms remain effective against known threats, they exhibit structural limitations when confronted with adaptive, multi-layered, and post-quantum adversarial models. To address these deficiencies, this study introduces CITADEL-DoH, a unified security architecture integrating quantum-resistant AI through hybrid LWE-CKKS encryption, decentralised trust enforcement via a proof-of-trust (PoT) protocol combined with federated byzantine agreement (FBA), and hardware-assisted integrity verification using FPGA-accelerated verifiable delay functions (VDFs). Novel contributions further include the application of topological data analysis (TDA) for encrypted anomaly detection and liquid time-constant (LTC) networks for adaptive traffic modelling. Experimental evaluation demonstrates 99.3% detection accuracy for domain generation algorithm (DGA) attacks and sub-millisecond query verification latency. Results confirm enhanced scalability, resilience to evasion, and robustness against emerging quantum-era threats.

Data centers rely on Priority-based Flow Control (PFC) to achieve lossless data transmission in Ethernet networks. To avoid buffer overflow, PFC pauses flows in a coarse-grained manner, which brings potential problems, e.g., Head-of-Line (HoL) blocking, and PFC deadlock. Although the state-of-the-art approach BFC with per-flow backpressure tackles some of the limitations of PFC, it faces implementation challenges due to the need for a large number of queues. In this paper, we present F-PFC, a fine-grained flow control scheme that only leverages a small amount of queues to address the limitations of PFC. Specifically, F-PFC first designs a fine-grained flow backpressure scheme to adjust the intensity of flow control adaptively. With different levels of flow backpressure, F-PFC ensures high throughput and low latency simultaneously. Then, F-PFC presents an accurate flow identification scheme to locate flows that really contribute to congestion. Finally, F-PFC presents a dynamic queue assignment and scheduling scheme to isolate congestion flows with limited queues. We theoretically analyze the performance of F-PFC and present the implementation of F-PFC. Extensive testbed experiments and large-scale simulations verify the performance of F-PFC. The experimental results show that F-PFC reduces tail latency by at least 33% and queue occupancy by 46% compared with state-of-the-art approaches.

Complex event recognition (CER) refers to identify-ing specific patterns composed of several primitive events in event stores. Since full-scanning event stores to identify primitive events that hold query constraint conditions incurs costly I/O overhead, a mainstream and practical approach is to use index techniques to obtain these events. However, prior index-based approaches suffer from significant I/O and sorting overhead when processing the query with high predicate selectivity or long query window, which leads to high query latency. To address this issue, we propose ACER, a Range Bitmap-based index, to accelerate CER. Firstly, ACER achieves a low index space overhead by grouping the events with the same type into a cluster and compressing the cluster data, reducing I/O overhead when reading indexes. Secondly, ACER builds Range Bitmaps for queried attributes and ensures that the events of each cluster in the index block are chronologically ordered. Then, ACER can always obtain ordered query results for a specific event type through merge operations, avoiding sorting overhead. Most importantly, ACER avoids unnecessary disk accesses in indexes and events via window-wise filtering, thus reducing the I/O overhead further. Lastly, we propose an enhanced version of ACER (ACER-E) by optimizing the read/write operation of index blocks and variable query order. Our extensive experiments demonstrate that ACER and ACER-E reduce the query latency by up to one order of magnitude compared with SOTA techniques.

Head orientation tracking has many potential applications in many fields,e.g., human-computer interaction, AR, and VR. In recent years, a large amount of existing work only focuses on the positioning of the user but ignore the tracking of the head orientation. Undoubtedly, with the information of the user's head orientation, many applications will have more opportunities for performance enhancement and better user experience. However, reviewing existing works regarding head tracking, the CV-based solutions have limited tracking angle range and privacy issues, the IMU-based solutions have accumulated errors, and the traditional microphone array-based solutions have low accuracy. Thus, none of these methods provides accurate and stable head orientation. In this paper, we propose EHeadTracker, an enhanced fine-grained head orientation tracking system based on head-mounted acoustic devices. This system recognizes rich head motions and achieves high-precision head direction tracking, while solving the problem of pivot point initialization. The experimental results show that the system can achieve an average error of 6 degrees in the head orientation tracking. To the best of our knowledge, EHeadTracker is the first system to use head-mounted acoustic devices to achieve head orientation tracking and has the highest accuracy in all current work.

Rotation is a fundamental form of motion and rotation speed measurement holds paramount importance for assessing the health and performance of machinery with rotating components. However, existing measurement systems often face challenges such as limited measurement distance, low accuracy, and complex installation or maintenance processes. In this paper, we propose RoLEX, a LoRa-based rotation speed measurement system for long-distance and contactless monitoring of rotating machinery in ubiquitous scenarios. RoLEX employs a novel Signal Selection method to eliminate chirp interference and adapt to varying rotation speeds, along with a Boost Sensing method to enhance sampling rates and an advanced feature processing algorithm for precise rotation speed estimation and tracking. Comprehensive experiments validate that RoLEX achieves a measurement distance of 50 m, approximately 17 times farther than the latest wireless rotation speed measurement systems. Moreover, RoLEX is robust to interference and obstructions (including through-wall scenarios) and achieves an average measurement error less than 0.69% across different rotation speeds (100 - 5100 Revolutions Per Minute). For tracking performance, RoLEX achieves a relative error less than 2.8% in 90% of cases. We also present a case study to highlight RoLEX's practical applicability in real-world scenarios.

Deploying large language models (LLMs) in mobile and edge computing environments is constrained by limited on-device resources, scarce wireless bandwidth, and frequent model evolution. Although edge-cloud collaborative inference with speculative decoding (SD) can reduce end-to-end latency by executing a lightweight draft model at the edge and verifying it with a cloud-side target model, existing frameworks fundamentally rely on tight coupling between the two models. Consequently, repeated model synchronization introduces excessive communication overhead, increasing end-to-end latency, and ultimately limiting the scalability of SD in edge environments. To address these limitations, we propose FlexSpec, a communication-efficient collaborative inference framework tailored for evolving edge-cloud systems. The core design of FlexSpec is a shared-backbone architecture that allows a single and static edge-side draft model to remain compatible with a large family of evolving cloud-side target models. By decoupling edge deployment from cloud-side model updates, FlexSpec eliminates the need for edge-side retraining or repeated model downloads, substantially reducing communication and maintenance costs. Furthermore, to accommodate time-varying wireless conditions and heterogeneous device constraints, we develop a channel-aware adaptive speculation mechanism that dynamically adjusts the speculative draft length based on real-time channel state information and device energy budgets. Extensive experiments demonstrate that FlexSpec achieves superior performance compared to conventional SD approaches in terms of inference efficiency.

Space Situational Awareness (SSA) relies on Low Earth Orbit (LEO) satellites to capture continuous, highresolution imagery critical for identifying space threats. The vast volume of SSA images overwhelms satellite network throughput, hindering timely transmission and processing. This paper presents a reinforcement learning (RL)-driven distributed sparse coding framework that integrates novel compression algorithms with orbital RL to address these challenges. First, we introduce an Aggregated Dictionary Learning (ADL) algorithm and a Context-aware Adaptive Binary Arithmetic Coding (CABAC) algorithm, achieving a 93.78% compression ratio by exploiting the high sparsity and spatiotemporal redundancy of SSA images. Second, the proposed compression workflow is deployed across LEO satellites in a distributed manner, where both overlapping and non-overlapping regions of images are dynamically partitioned and processed in parallel to optimize resource utilization and reduce latency. An Orbital Double Deep Q-Network (DQN) framework is proposed to optimize task offloading decisions by (1) integrating orbital dynamics into the state space, and (2) adaptively partitioning images based on visible LEO resources. Evaluations demonstrate that our framework achieves 100% task completion under visibility constraints and a 51.61% reduction on CPU and RAM occupation time compared to centralized processing.

Robotic swarm coordination holds transformative potential for applications such as warehouse automation, search & rescue, and entertainment. However, approaches relying on wearable devices or vision-based systems are often constrained by hardware-intensive, high computational requirements, reliance on line-of-sight, and privacy concerns. Wireless sensing, particularly using Channel State Information (CSI), offers a promising alternative by translating environmental perturbations into CSI variation data. Nevertheless, existing CSI-based systems face significant challenges in domain adaptation, resource limitation, and scalability issues. This paper introduces MaestroBot, a hierarchical motion coordination system that combines distributed CSI-based wireless sensing with domain-adaptive learning to address these limitations. For leader robots, the system features a lightweight hand gesture recognition model, built on a “Hybrid-Single” knowledge distillation framework, achieving up to 95.87% accuracy while maintaining adaptability across diverse domains. For follower robots, the hierarchical motion propagation model leverages localized CSI analysis and dual-layer error correction mechanisms to deliver 97.2% accuracy with a low latency of 0.085 seconds, even in multi-row formations. Additionally, its cost-effective hardware design ensures practical scalability and real-world deployability. These results position MaestroBot as an efficient, robust, and privacy-preserving solution for large-scale robotic swarm coordination in dynamic environments.

Today's mobile applications oftentimes need to detect user-defined complex patterns (e.g., the mysterious “phantom traffic jam”) over data streams to support decision making. It is achieved by continuously creating candidate instances that have partially matched a pattern, and meanwhile aggregating common instances (across patterns) for efficiency enhancement. Existing aggregation approaches are taken in a straightforward or intuitive manner, incurring an exponential solution space and thus having to be executed offline. This paper explores how to significantly accelerate aggregation so as to make pattern detection online executable, even suited to the emerging serverless runtime that involves complicated state synchronizations among distributed cloud functions. By comprehensively investigating a wide variety of mobile data streams, we note the existence of a latent hierarchical cluster structure among complex patterns (in terms of their instance similarities), which can be utilized to quickly aggregate common instances without going through the exponential solution space. To extract the latent information, we devise a content-aware structural entropy minimization algorithm to properly determine intra-cluster patterns, together with a lightweight differential compensation mechanism to maintain those inter-cluster “residual” relations among patterns. Evaluations on real-world vehicle and sensor network data streams illustrate that the resulting approach, dubbed PatternInsight, saves the aggregation time by 10× to 50× and reduces the instance size by 40%.

This paper addresses theJoint deployment ofUnmanned aerial vehicles (UAVs) and surveillance caMeras withPath planning (JUMP), aiming to deploy a fixed number of UAVs and budget-limited surveillance cameras, to achieve continuous omnidirectional monitoring. Specifically, the objective is to maximize the monitoring durations of target objects in each of all horizontal directions within a given task duration. We propose an approach for JUMP, which is proved to be NP-hard. Our approach achieves a$\frac{1}{6}-\varepsilon _{1}$approximation ratio in general and$\frac{1}{4}-\varepsilon _{1}$when camera costs are uniform. Firstly, we introduce spatio-temporal discretization to approximate JUMP. Secondly, we partition the solution space of JUMP from spatial perspective and refine the space by addressing a variant of obstacle-avoiding shortest path problem spatio-temporal perspective. Thirdly, we reformulate the problem as a classical problem of Monotone Submodular set function Maximization with one partition Matroid and two Knapsack constraints (MSMMK). To address MSMMK, we propose a$\frac{1}{6(1+\varepsilon )}$approximation algorithm, which outperforms the state-of-the-art with the same time complexity in general. Specifically, our algorithm achieves a$\frac{1}{4(1+\varepsilon )}$approximation ratio for special cases. Simulation results demonstrate our proposed approach outperforms five benchmark algorithms, yielding enhancements 13%-1446%. Moreover, field experiment results indicate that our approach surpasses comparison algorithms, achieving enhancements 23%-265%.

Mobile video analytics systems are often deployed in diverse environments, where domain shifts place higher demands on the responsiveness of adaptation for ”expert models” running on resource-constrained mobile devices. However, due to the lack of effective automated on-device model retrieval strategies and corresponding mobile adaptation mechanisms, most existing frameworks still rely on cloud-centric adaptation architectures rather than mobile-centric ones, leading to delayed responses to such domain shifts. We introduce a hierarchical mobile-cloud collaborative adaptation framework, AutoMOCHA, for continuous mobile video analytics in dynamically evolving environments. It includes three main contributions: (1) It facilitates efficient history expert model retrieval, without relying on dataset-based model validation, through an automatically constructed model taxonomy and indexing mechanism that does not require prior domain knowledge; (2) It proactively coordinates onboard lightweight model reuse and finetuning strategy with remote model retrieval and retraining to achieve responsive model adaptation with low latency; (3) It accelerates onboard model reuse and finetuning through designing a dedicated mobile model cache management strategy and sparse updating policy. Extensive evaluations on real-world video data across three DNN tasks demonstrate that AutoMOCHA improves model accuracy during adaptation by up to 3.4%, increases the post-clustering silhouette coefficient by up to 33%, and reduces retraining frequency by up to 10%, all while maintaining optimal response latency and retraining time.

Existing DNS over HTTPS defenses have demonstrated limited resilience against polymorphic traffic shaping, staged tunneling, and adaptive mimicry, largely because they rely on static learning pipelines and rigid cryptographic configurations. MIRAGE-DoH was designed to examine whether adaptive inference, persistent structural encoding, and calibrated cryptographic agility could be integrated into a deployable and measurable encrypted DNS control architecture. The framework combined flow-level Cognitive MetaAgents capable of internal reconfiguration, Topological Memory Networks that preserved stable geometric irregularities across temporal windows, and Metamorphic Cryptographic Shards that adjusted key encapsulation policies according to empirically calibrated threat severity. A Causal Counterfactual Environment modeled constrained attacker decision pathways, while Spectral Game Intelligence analyzed flow interaction graphs to anticipate structural attack transitions.Evaluation on extended CIC-DoH2023 and Gen-C-DDD-2022 datasets was conducted under fixed flow-level decision intervals with explicit accounting for synchronization overhead, spectral graph construction cost, and cryptographic rotation latency. Cross-dataset experiments yielded a mean detection accuracy of 97.8% with a 0.41% false positive rate, sustaining median inference latency of 62μs and cryptographic morph latency of 3.7 ms under load. Quantum-assisted inference was assessed through bounded simulations, indicating constrained information gain within the adopted lattice-based configuration, without asserting unconditional post-quantum immunity. These results demonstrated that adaptive encrypted DNS governance can be empirically grounded, operationally bounded, and stress-evaluated without reliance on unqualified claims of perfect security.

Node-centric Data Centers (NDCs) are highly flexible, cost-efficient, and failure-resilient, and have gained growing popularity in recent years. However, RDMA technology used in NDC still faces challenges, including high retransmission overhead, Head-of-Line Blocking (HoLB) and deadlock problems. Existing solutions for traditional data centers cannot simultaneously address these issues due to the unique topology and server transmission characteristics of NDC. In this paper, we propose a per-port flow control named PortFC for NDC. PortFC addresses the above problems through the designs of a Pause/Resume control signal, a per-port queue allocation method, an egress-detecting per-port flow control mechanism, and a server-aware queue scheduling method. Our evaluation shows that PortFC is free from retransmission, capable of eliminating HoLB and avoiding deadlocks. PortFC achieves 1.7-8.0 times higher throughput and reduces latency by 11.7%-87.7% compared to the state-of-the-art lossy RDMA based on IRN and the lossless RDMA method based on PFC. In particular, PortFC still demonstrates good performance in a Rail-only NDC with heterogeneous bandwidth domains.

High-definition (HD) cameras for surveillance and road traffic have experienced tremendous growth, demanding intensive computation resources for real-time analytics. Recently, offloading frames from the front-end device to the back-end edge server has shown great promise. In multi-stream competitive environments, efficient bandwidth management and proper scheduling are crucial to ensure both high inference accuracy and high throughput. To achieve this goal, we propose BiSwift, a bi-level framework that scales the concurrent real-time video analytics by a novel adaptive hybrid codec integrated with multi-level pipelines, and a global bandwidth controller for multiple video streams. The lower-level front-back-end collaborative mechanism (called adaptive hybrid codec) locally optimizes the accuracy and accelerates end-to-end video analytics for a single stream. The upper-level scheduler aims to accuracy fairness among multiple streams via the global bandwidth controller. The evaluation of BiSwift shows that BiSwift is able to real-time object detection on 9 streams with an edge device only equipped with an NVIDIA RTX3070 (8G) GPU. BiSwift improves 10%~21% accuracy and presents$1.2\sim 9\times $throughput compared with the state-of-the-art video analytics pipelines.

In modern mobile and edge applications, users and devices frequently encounter dynamic contexts, necessitating ondevice continual learning (CL) to ensure consistent performance of machine learning models. While existing research predominantly focused on developing lightweight CL frameworks, we identify that data scarcity is a more severe bottleneck for ondevice CL. In this work, we explore the potential of leveraging abundant cloud-side data to enrich scarce on-device data, and propose Delta, a private, efficient and effective data enrichment framework. Specifically, Delta introduces a directory dataset to decompose the data enrichment problem into device-side and cloud-side sub-problems without sharing raw sensitive data, mitigating privacy concerns and reducing communication overhead. Next, Delta designs a soft data matching strategy to effectively solve device-side sub-problem with sparse user data and an optimal data sampling scheme for cloud server to efficiently retrieve the most suitable dataset for enrichment. Further, Delta refines the data sampling scheme by jointly considering the impact of enriched data on both new and past contexts, mitigating the catastrophic forgetting issue from a new aspect. Finally, we enhance Delta’s applicability to more private and rare user contexts by integrating differential privacy and domain adaptation techniques. Comprehensive experiments across four typical edge computing tasks with varied data modalities demonstrate that Delta significantly enhances overall model accuracy by an average of 15.1%, 12.4%, 1.1% and 5.6% for visual, IMU, audio and textual tasks compared with few-shot CL. Additionally, Delta reduces communication costs by over 90% compared to federated CL, making it suitable for resource-constrained edge networks.

Distributed AI model training generates bursty, low-entropy elephant flows that challenge existing single-path transport protocols in multi-stage Clos networks, leading to congestion and inefficiency. Multipath transport emerges as a promising solution, leveraging multiple paths to balance traffic and enhance resilience. However, current multipath RDMA solutions suffer from scalability, congestion control, and load-balancing inefficiencies. This paper introduces Unified Delay-driven Multipath Protocol (UDMP), a novel approach that co-designs congestion control and load balancing using network delay as a unified signal. UDMP employs delay-gradient-based congestion control to precisely resolve unavoidable congestion. Moreover, UDMP leverages delay-assisted load balancing to shift traffic across paths with minimal latency adaptively, maintaining throughput when encountering avoidable congestion. A novel Token Pool design integrates these components, eliminating per-path state overhead while achieving fine-grained traffic distribution. Implementations on DPDK and NS3 demonstrate that UDMP achieves up to 2x higher throughput and reduces flow completion times by up to 30% compared to state-of-the-art methods like MPRDMA and QP-Scaling. These results highlight UDMP’s effectiveness in meeting the stringent performance requirements of modern distributed AI training workloads.

In modern GPU-centric clusters, large-scale AI training relies on two distinct communication domains: a high-bandwidth intra-node domain using proprietary interconnects (e.g., NVLink), and a scale-out inter-node network domain (e.g., RDMA). We observe that the widely-used ring algorithm, often create a significant load imbalance across these domains. This leads to the counter-intuitive scenario where the expensive, high-bandwidth intra-node domain becomes a performance bottleneck, while the inter-node network remains underutilized. This inefficiency is further exacerbated by the disparity in bandwidth provisioning: inter-node network bandwidth is generally more cost-effective and accessible, whereas intra-node bandwidth is often proprietary and more costly to scale. To address this fundamental imbalance, we propose RAIL, aimed at resolving the intra-node bottleneck by strategically rearranging inter-GPU communication paths. This rebalancing ensures that traffic loads are appropriately matched with the distinct transmission capabilities of each domain, thereby maximizing overall communication performance. RAIL incorporates a Load Distributing Strategy (LDS) that can accurately partition physical nodes into logical nodes based on the a transmission capabilities of both domains, shifting excess traffic from the overloaded intra-node domain to the underutilized network domain. Additionally, the Intra-Rail Strategy (IRS) leverages topological characteristics to ensure optimal communication paths through the network domain between logical nodes. Our evaluation demonstrates that RAIL effectively mitigates congestion and achieves a 30.7% average increase in collective communication bus bandwidth compared to the widely-used NCCL solution.

This paper proposes the concept of a “loss-tolerant” RDMA network, instantiating as NüWa. It reveals the fundamental issues under a lossy fabric — packet losses and repetitive retransmission timeouts (RTOs) cause severe performance degradation and even service interruption. The loss-tolerant RDMA must avoid “important” packet losses that trigger RTOs. However, existing loss-protection mechanisms fail to identify these packets precisely. They either generate massive misprotection or ignore selective repeat loss recovery, resulting in buffer overflows and failures of RTO protection. To tackle these issues, NüWa thoroughly analyzes distinct loss-recovery schemes and RTO reasons for commodity NICs. It designsswitch modeandNIC modeto accurately identify and protect all important packets. The switch mode inherently supports the widely deployed non-programmable NICs, while the NIC mode offloads identification complexity to advanced programmable NICs. With effective RTO avoidance, it improves flow completion time (FCT) by 2 ∼ 10× compared to state-of-the-art (SOTA) solutions. In severe incast and large-scale networks, it reduces FCT by 100× compared with a lossless fabric. In storage applications, N¨uWa improves IOPS by ∼ 300% compared to vanilla lossy fabric. For typical AI Workloads, it accelerates AllReduce/AlltoAll communication by 5.5 ∼ 13.7× compared to SOTA lossy network solutions.

In modern datacenters, the effectiveness of end-to-end congestion control (CC) is quickly diminishing with the rapid bandwidth evolution. Per-hop flow control (FC) can react to congestion more promptly. However, a coarse-grained FC can result in Head-Of-Line (HOL) blocking. A fine-grained, per-flow FC can eliminate HOL blocking caused by flow control, however, it does not scale well. This paper presents Pyrrha, a scalable flow control approach that provably eliminates HOL blocking while using a minimum number of queues. In Pyrrha, flow control first takes effect on the root of the congestion, i.e., the port where congestion occurs. And then flows are controlled according to their contributed congestion roots. A prototype of Pyrrha is implemented on Tofino2 switches. Compared with state-of-the-art approaches, the average FCT of uncongested flows is reduced by 42%-98%, and 99th-tail latency can be$1.6\times $-$215\times $lower, without compromising the performance of congested flows.

Network failure uncertainty and traffic demand uncertainty pose a series of challenges to efficiently solve network traffic engineering (TE) problems in the production wide-area networks (WANs). Prior works especially predict-then-optimize paradigm in TE systems encounter a significant mismatch between the upstream prediction task and the downstream optimization task, leading to potential demand loss risk and further degrading the performance of customer-facing applications in the cloud. In this paper, we present TUFTTE, a risk-driven TE optimization framework, where the objective function and constraints can be flexibly specified by the network operators. We employ a differentiable optimization layer to guide the prediction model with the objective of minimizing the network risks. When the network risk is defined as the demand loss, our approach can provably converge to the global optimum. Experimental results on real-world topology demonstrate that TUFTTE reduces the demand loss by 11.59% on average compared to the state-of-the- art algorithm. We believe that our paper deeply explores the potential of prediction and promotes the understanding on how to develop and deploy a general learning module into the commercial TE optimization.

Considering limited on-device resources, current practices are attempting to deploy a system-level mixture-of-experts (MoE)-based foundation LLM on a mobile device to serve multiple apps and support mobile intelligence. However, mobile apps are hard to customize their services that require fine-tuning adapters associated with the LLM using private in-app data. The difficulty arises due to both the limited on-device resources and the restricted control that apps have over the foundation LLM. To address this issue, in this work, we propose LiteMoE, a novel proxy submodel tuning framework that supports mobile apps to efficiently fine-tune customized adapters on devices using proxy submodels. The key technique behind LiteMoE is a post-training submodel extraction method, whereby without additional retraining, we can identify and reserve critical experts, match and merge moderate experts, to extract a lightweight and effective proxy submodel from the foundation LLM for a specific app. To further enhance scalability and adaptability, LiteMoE incorporates adapter reuse and continuous tuning mechanisms to handle multi-task requirements and evolving user preferences. We implemented a prototype of LiteMoE and evaluated it over various MoE-based LLMs and mobile computing tasks. The results show that with LiteMoE, mobile apps are able to fine-tune customized adapters on resource-limited devices, achieving 12.7% accuracy improvement and 6.6× memory reduction compared with operating the original foundation LLM.

Edge-end collaborative learning trains models with exchanged knowledge through distributed interaction, alleviating the cloud's burden. Edge-end heterogeneous collaborative learning further enables edge servers and end devices to train models of different scales in parallel based on their computational capabilities. This technology supports various applications in different resource conditions and improves server resource utilization. However, implementing it is challenging due to heavy communication costs and high global costs (time and energy). To this end, this paper proposes a novel prototype-based edge-end heterogeneous collaborative learning method and an optimization algorithm, which improves model performance and reduces training costs. We first use prototypes to perform collaborative learning and analyze the convergence. Prototypes are computed as mean feature vectors from different classes. The aggregated prototypes help capture class information on end devices and generate data on edge servers. Then, we study how to determine prototype selection and edge association to minimize training time, energy consumption, and prototype approximation error under a limited reward budget, which is proven to be NP-hard. We split the original problem into two subproblems. The first is solved in the closed form. Through approximation and reformulation, the second is transformed into a submodular maximization problem with knapsack and matroid constraints. On this basis, we propose an approximation algorithm with a theoretical guarantee. Finally, by simulation and field experiments, our method takes 3.61% of communication costs to improve heterogeneous edge and end models' accuracy by at least 5.16% and 2.77% compared with five baselines. The proposed algorithm outperforms others by at least 9.78% in terms of global cost.

To guide consumers to mobile livestreams that involve their interested products, it is necessary to have a good understanding of livestreaming contents, and the key task is to accurately recognize the products being promoted by streamers. The mainstream cloud-based service framework is challenged by the high concurrency of requests, the high overhead of multimodal recognition, and the requirement of low latency. To break these bottlenecks, we propose a new device-cloud collaborative service framework, where each streamer's mobile device holds unimodal model that can process most of video frames and also uploads the extracted unimodal features to facilitate the cloud-side multimodal recognition of the remaining few frames. On-device unimodal model is further incrementally trained over the samples constructed by leveraging the streamers' manual labeling behaviors, leading to personalized version that can adapt to the heterogeneous and dynamic livestreaming contents of different streamers. Nevertheless, the device-side personalized unimodal features are misaligned in feature space and cannot be directly fused into the cloud-side multimodal model. We thus design a pluggable prompt generation module to transform the personalized unimodal features into prompt embeddings and prepend them to the original inputs of the multimodal backbone network, instructing feature fusion in a streamer-specific manner. We extensively evaluate on the public Fashion-Gen dataset and an industrial dataset collected from Taobao Live. We also perform online testing and practical overhead testing in Taobao Live. Evaluation results reveal the effectiveness and efficiency of our design as well as the consistent advantage over existing baselines.

The emergence of long-context text applications utilizing large language models (LLMs) has presented significant scalability challenges, particularly in memory footprint. The linear growth of the Key-Value (KV) cache, which stores attention keys and values to reduce redundant computations, can significantly increase memory usage and may prevent models from functioning properly in memory-constrained environments. To address this issue, we propose a novel approach called Cache Sparse Representation (CSR), which converts the KV cache by transforming the dense Key-Value cache tensor into sparse indexes and weights, offering a more memory-efficient representation during LLM inference. Furthermore, we introduce NeuralDict, a novel neural network-based method to automatically generate the dictionary used in our sparse representation. Our extensive experiments demonstrate that CSR matches the performance of state-of-the-art KV cache quantization algorithms while ensuring robust functionality in memory-constrained environments.

Junyi Chen, Shihao Bai, Zaijun Wang, Siyu Wu, Chuheng Du, Hailong Yang, Ruihao Gong, Shengzhong Liu, Fan Wu, Guihai Chen. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Recent advancements in deep learning have significantly increased AI processors' energy consumption, which is becoming a critical factor limiting AI development. Dynamic Voltage and Frequency Scaling (DVFS) stands as a key method in power optimization. However, due to the latency of DVFS control in AI processors, previous works typically apply DVFS control at the granularity of a program's entire duration or sub-phases, rather than at the level of AI operators.

With the rise of deep learning, many companies have developed domain-specific architectures (DSAs) optimized for AI workloads, with Ascend being a representative. To fully realize the operator performance on Ascend, effective analysis and optimization is urgently needed. Compared to GPU, Ascend requires users to manage operations manually, leading to complex performance issues that require precise analysis. However, existing roofline models face challenges of visualization complexity and inaccurate performance assessment. To address these needs, we introduce a component-based roofline model that abstracts components to capture operator performance, thereby effectively identifying bottleneck components. Furthermore, through practical operator optimization case studies, we illustrate a comprehensive process of optimization based on roofline analysis, summarizing common performance issues and optimization strategies. Finally, extensive end-to-end optimization experiments demonstrate significant model speed improvements, ranging from 1.07× to 2.15×, along with valuable insights from practice.

Existing work on large language model (LLM) personalization assigned different responding roles to LLMs, but overlooked the diversity of queriers. In this work, we propose a new form of querier-aware LLM personalization, generating different responses even for the same query from different queriers. We design a dual-tower model architecture with a cross-querier general encoder and a querier-specific encoder. We further apply contrastive learning with multi-view augmentation, pulling close the dialogue representations of the same querier, while pulling apart those of different queriers. To mitigate the impact of query diversity on querier-contrastive learning, we cluster the dialogues based on query similarity and restrict the scope of contrastive learning within each cluster. To address the lack of datasets designed for querier-aware personalization, we also build a multi-querier dataset from English and Chinese scripts, as well as WeChat records, called MQDialog, containing 173 queriers and 12 responders. Extensive evaluations demonstrate that our design significantly improves the quality of personalized response generation, achieving relative improvement of 8.4% to 48.7% in ROUGE-L scores and winning rates ranging from 54% to 82% compared with various baseline methods.

Multi-tenancy in modern cloud service colocates multiple virtual machines (VMs) into physical machines (PMs) to improve resource efficiency. However, co-location introduces interference among VMs, potentially degrading the quality-ofservice (QoS) for users. Previous methods predict QoS degradation and schedule VMs accordingly, but they often overlook important information provided by VM metrics and are hard to integrate with real-world cloud schedulers. Considering the above factors, we present TRACE, a novel QoS-aware, lightweight, and decoupled recommender for VM scheduling. Firstly, TRACE employs a dual-tower feature extraction mechanism that independently extracts metrics from VMs and PMs, thereby reducing the time complexity of the model. Secondly, the dual-tower is enhanced by Deep & Cross Networks to explicitly model cross-feature interactions, and we further incorporate a Set Transformer to process overlooked multi-VM metrics from the PM. Thirdly, TRACE designs a trainable similarity gate and an adaptive mask to filter suboptimal migrations, decoupling it from the scheduler for easy integration. Experimental results on data collected from real-world clusters show that TRACE outperforms state-of-the-art methods in QoS prediction accuracy and ranking quality, achieving at least 6.3 % QoS improvements.

Diffusion models have achieved remarkable progress in text-to-image generation, enabling the rise of personalized models. A key challenge in personalized generation is to provide users with precise control while ensuring high fidelity to the content. To address this, we introduce VariGen: a framework that empowers users to achieve fine-grained, layout-controllable personalized image generation. VariGen employs the Variational Detail-Aware Feature Extractor to capture intricate details from reference subjects and the Dual Layout Control Mechanism to integrate layout specifications seamlessly into the generation process. We demonstrate that VariGen achieves superior performance through extensive experimentation, offering unparalleled creative freedom and fidelity. To our knowledge, this is the first work to enable users to “create anything, anywhere” with such precision and flexibility.

Cloud providers require high-throughput traffic to test the effectiveness of congestion control (CC) configurations (i.e., CC algorithm selection and their parameter settings) in networks. A network tester capable of evaluating CC configurations needs to fulfill the following requirements: (R1) Capable of generating traffic with CC behaviors. (R2) Ability to customize CC algorithms. (R3) High throughput CC traffic generation. However, existing network testers fail to meet these requirements simultaneously. The paper presents Marlin, a novel high-throughput network tester designed for CC evaluation. Marlin leverages a high-throughput, low-programmability device to amplify the traffic generated by a low-throughput, high-programmability device. The low-throughput device is responsible for complex computational tasks, such as running CC and flow scheduling algorithms, and communicates with the high-throughput device at a high frequency using small packets to instruct it to generate high-throughput traffic with CC behaviors. This hybrid approach allows for customizable, high-throughput CC testing. Our experiments demonstrate that Marlin can accurately emulate CC behaviors and replicate real-world scenarios. Marlin can generate 1.2 Tbps of CC traffic using a single programmable switch pipeline and one 100 Gbps port of an FPGA NIC, supporting up to 65,536 concurrent flows.

Large Multimodal Models (LMMs) have shown significant progress in various complex vision tasks with the solid linguistic and reasoning capacity inherited from large language models (LMMs). Low-rank adaptation (LoRA) offers a promising method to integrate external knowledge into LMMs, compensating for their limitations on domain-specific tasks. However, the existing LoRA model serving is excessively computationally expensive and causes extremely high latency. In this paper, we present an end-to-end solution that empowers diverse vision tasks and enriches vision applications with LoRA LMMs. Our system, VaLoRA, enables accurate and efficient vision tasks by 1) an accuracy-aware LoRA adapter generation approach that generates LoRA adapters rich in domain-specific knowledge to meet application-specific accuracy requirements, 2) an adaptive-tiling LoRA adapters batching operator that efficiently computes concurrent heterogeneous LoRA adapters, and 3) a flexible LoRA adapter orchestration mechanism that manages application requests and LoRA adapters to achieve the lowest average response latency. We prototype VaLoRA on five popular vision tasks on three LMMs. Experiment results reveal that VaLoRA improves 24-62% of the accuracy compared to the original LMMs and reduces 20-89% of the latency compared to the state-of-the-art LoRA model serving systems.

In data center networks, various types of traffic with strict performance requirements operate simultaneously, necessitating effective isolation and scheduling through priority queues. However, most switches support only around ten priority queues. Virtual priority can address this limitation by emulating multi-priority queues on a single physical queue, but existing solutions often require complex switch-level scheduling and hardware changes. Our key insight is that virtual priority can be achieved by carefully managing bandwidth contention in a physical queue, which is traditionally handled by congestion control (CC) algorithms. Hence, the virtual priority mechanism needs to be tightly coupled with CC. In this paper, we propose PrioPlus, a CC enhancement algorithm that can be integrated with existing congestion control schemes to enable virtual priority transmission. PrioPlus assigns specific delay ranges to different priority levels, ensuring that flows transmit only when the delay is within the assigned range, effectively meeting virtual priority requirements. Compared to Swift CC with physical priority queues, PrioPlus provides strict priority for high-priority flows without impacting performance sensibly. Meanwhile, it benefits low-priority flows from 25% to 41% as its priority-aware design enhances CC's ability to fully utilize available bandwidth once higher-priority traffic completes. As a result, in coflow and model training scenarios, PrioPlus improves job completion times by 21% and 33%, respectively, compared to Swift with physical priority queues.

This paper optimizes the computation efficiency for DNN-based video analytics on interconnected edge devices like surveillance cameras and unmanned aerial vehicles (UAVs). Existing solutions depend heavily on computation offloading to edge servers but overlook the potential for cross-device collaboration and leave resources on some edge devices underutilized. We instead propose AdaCollab, an adaptive multi-device collaboration framework for resource-efficient distributed edge video analytics. On the one hand, as the machine perception complexity fluctuates with runtime video content, AdaCollab dynamically calibrates DNN inspection configurations on each device without degrading the model accuracy. On the other hand, AdaCollab aligns mismatched resources and workload among edge devices by selectively offloading partial computations from overloaded devices to underloaded devices, optimizing both computation and bandwidth resource utilization. Through extensive evaluations with large-scale real-world surveillance videos on testbeds of heterogeneous NVIDIA Jetson platforms, AdaCollab outperforms the SOTA baselines by up to 25.8% in frame processing through-put and 19.8% in DNN accuracy, while exhibiting enhanced robustness under restricted network bandwidth.

Optimizing anycast catchment to ensure clients connect to their expected PoPs remains a significant challenge in production networks, primarily due to BGP’s inherent lack of performance awareness. Existing approaches either lack fine-grained routing policy control or prove operationally impractical. In this paper, we present AnyTuner, an automatic route management system that strategically optimizes anycast performance through AS-path prepending and BGP communities. AnyTuner integrates three key components: a measurement-driven optimization loop, a neural network for PoP traffic matrix prediction, and an automated recommendation engine for configuration generation. Deployed for over two years on a commercial cloud platform, AnyTuner demonstrates remarkable efficacy in our evaluation across 18 PoPs and 9,000+ probes, achieving a 39.9% reduction in median RTT (from 36.3 ms to 21.8 ms) compared to policy-free baseline configurations and outperforming state-of-the-art alternatives like AnyOpt by 27.8%.

BCube is a modular data center network.Compared with other topologies, BCube has natural advantages, such as lower deployment costs and stronger failure recovery capabilities.However, RDMA technology used in BCube still faces challenges, including high retransmission overhead, Head-of-Line Blocking (HoLB) and deadlock problems.Existing solutions for traditional data centers cannot simultaneously address these issues due to the unique topology and server transmission characteristics of BCube.In this paper, we propose a per-port flow control named PortFC for BCube.PortFC addresses the above problems through the designs of a Pause/Resume control signal, a per-port queue allocation method, an egress-detecting per-port flow control mechanism, and a serveraware queue scheduling method.Our evaluation shows that PortFC is free from retransmission, capable of eliminating HoLB and avoiding deadlocks.PortFC achieves 1.7-8.0times higher throughput and reduces latency by 11.7%-87.7%compared to the state-of-the-art

Network failure uncertainty and traffic demand uncertainty pose a series of challenges to efficiently solve network traffic engineering (TE) problems in the production wide-area networks (WANs). Prior works especially predict-then-optimize paradigm in TE systems encounter a significant mismatch between the upstream prediction task and the downstream optimization task, leading to potential demand loss risk and further degrading the quality of services in the cloud. In this paper, we present TUFTTE, a risk-driven TE optimization framework, where the objective function and constraints can be flexibly specified by the network operators. We employ a differentiable optimization layer to guide the prediction model with the objective of minimizing the network risks. When the network risk is defined as the demand loss, our approach can provably converge to the global optimum. Experimental results on real-world topology demonstrate that TUFTTE reduces the demand loss by 11.59 % on average compared to the state-of-the-art algorithm. We believe that our paper deeply explores the potential of prediction and promotes the understanding on how to develop and deploy a general learning module into the commercial TE optimization.

In many practical natural language applications, user data are highly sensitive, requiring anonymous uploads of text data from mobile devices to the cloud without user identifiers. However, the absence of user identifiers restricts the ability of cloud-based language models to provide personalized services, which are essential for catering to diverse user needs. The trivial method of replacing an explicit user identifier with a static user embedding as model input still compromises data anonymization. In this work, we propose to let each mobile device maintain a user-specific distribution to dynamically generate user embeddings, thereby breaking the one-to-one mapping between an embedding and a specific user. We further theoretically demonstrate that to prevent the cloud from tracking users via uploaded embeddings, the local distributions of different users should either be derived from a linearly dependent space to avoid identifiability or be close to each other to prevent accurate attribution. Evaluation on both public and industrial datasets using different language models reveals a remarkable improvement in accuracy from incorporating anonymous user embeddings, while preserving real-time inference requirement.

In the field of auction design, leveraging deep learning to solve optimal auctions from sampled data has become a promising direction. However, real-world contexts often involve uncertain data, which would severely affect the auction performance, but it is lacking consideration in existing works. To address this challenge, we incorporate these uncertainties into auction design metrics, and frame this challenge as a robust data-driven auction design problem. To solve this problem, we first propose the GAT method, where we introduce the process of problem relaxation and transformation to address the non-differentiable variable presented in the original problem, and further propose an adversarial training algorithm to solve the mini-max problem after transformation. Moreover, to obtain moderately robust auctions, we propose two methods to select the robust coefficient, which provides guidance and insights for selecting robust auctions based on generalization and performance metrics. Finally, with the insights from the GAT method, we further propose the SAT method, where we employ a strict and unified IC constraint that extends from the GAT method, which provides strong IC guarantees and stable revenue in uncertain environments. Experiments on both constructed and real-world datasets show that our robust methods effectively improve the performance of auctions in terms of revenue and IC guarantees.

Low-latency interactive video streaming services critically depend on robust congestion control algorithms (CCA). However, existing CCAs often exhibit frequent self-induced oscillations from unconstrained probing, causing periodic heavy queuing and degrading the user experience. We propose Themis, a novel end-to-end CCA designed to achieve stabilized near-zero queuing delay with high bitrate. By precisely quantifying the trade-off between bitrate and queuing delay within a utility feedback mechanism, Themis effectively controls the amplitude and frequency of probing while ensuring fairness. In parallel, Themis incorporates the awareness of the queuing state through an adaptive-pacing method, combined with utility feedback to guide a three-phase bitrate adjustment strategy. This enables rapid and stable convergence to the optimal utility. We implemented Themis in QUIC, evaluated it on the Mahimahi and conducted a 90-day large-scale A/B test in a real-world network. Compared to state-of-the-art CCAs, Themis effectively suppresses self-induced oscillations, increases the average frame bitrate by 66.8%, and reduces the average frame delay by 13.5%, demonstrating a superior trade-off between high bitrate and low queuing delay.

With the rapid growth of Low Earth Orbit (LEO) satellite networks, satellite-IoT systems using the LoRa technique have been increasingly deployed to provide widespread Internet services to low-power and low-cost ground devices. However, the long transmission distance and adverse environments from IoT satellites to ground devices pose a huge challenge to link reliability, as evidenced by the measurement results based on our real-world setup. In this paper, we propose a blind coherent combining design named B2LoRa to boost LoRa transmission performance. The intuition behind B2LoRa is to leverage the repeated broadcasting mechanism inherent in satellite-IoT systems to achieve coherent combining under the low-power and low-cost constraints, where each re-transmission at different times is regarded as the same packet transmitted from different antenna elements within an antenna array. Then, the problem is translated into aligning these packets at a fine granularity despite the time, frequency, and phase offsets between packets in the case of frequent packet loss. To overcome this challenge, we present three designs — joint packet sniffing, frequency shift alignment, and phase drift mitigation to deal with ultra-low SNRs and Doppler shifts featured in satellite-IoT systems, respectively. Finally, experiment results based on our real-world deployments demonstrate the high efficiency of B2LoRa.

In creator economy, the online platform posts contract to incentivize creators for high-quality content. The creators can use human effort or Generative Artificial Intelligence (GenAI) to assist them to create content for rewards. In this work, we identify that the existing contract suffers from GenAI-Induced Effort Degradation, where creators may not try their best to create high-quality content, but instead turn to GenAI to save effort and extract additional reward. Our results show that GenAI-induced effort degradation arises from failing to provide the same profit for the creators generating the same quality content. With this observation, we design the contract for creator economy in the era of GenAI as a classical contract for human content and a reward for the GenAI content which depends on the estimation of the human effort cost required to achieve GenAI's content quality. To overcome the curse of dimensionality in learning the optimal contract that maximizes platform profit with a high-quality content guarantee, we design an adaptive action pruning method that gradually filters out contracts resulting in low quality. Simulation experiments on synthetic data demonstrate the effectiveness of our approach.

Small language models (SLMs) support efficient deployments on resource-constrained edge devices, but their limited capacity compromises inference performance. Retrieval-augmented generation (RAG) is a promising solution to enhance model performance by integrating external databases, without requiring intensive on-device model retraining. However, large-scale public databases and user-specific private contextual documents are typically located on the cloud and the device, respectively, while existing RAG implementations are primarily centralized. To bridge this gap, we propose DRAGON, a distributed RAG framework to enhance on-device SLMs through both general and personal knowledge without the risk of leaking document privacy. Specifically, DRAGON decomposes multi-document RAG into multiple parallel token generation processes performed independently and locally on the cloud and the device, and employs a newly designed Speculative Aggregation, a dual-side speculative algorithm to avoid frequent output synchronization between the cloud and device. A new scheduling algorithm is further introduced to identify the optimal aggregation side based on real-time network conditions. Evaluations on real-world hardware testbed demonstrate a significant performance improvement of DRAGON—up to 1.9X greater gains over standalone SLM compared to the centralized RAG, substantial reduction in per-token latency, and negligible Time to First Token (TTFT) overhead.

Mobile agents rely on Large Language Models (LLMs) to plan and execute tasks on smartphone user interfaces (UIs). While cloud-based LLMs achieve high task accuracy, they require uploading the full UI state at every step, exposing unnecessary and often irrelevant information. In contrast, local LLMs avoid UI uploads but suffer from limited capacity, resulting in lower task success rates. We propose $\textbf{CORE}$, a $\textbf{CO}$llaborative framework that combines the strengths of cloud and local LLMs to $\textbf{R}$educe UI $\textbf{E}$xposure, while maintaining task accuracy for mobile agents. CORE comprises three key components: (1) $\textbf{Layout-aware block partitioning}$, which groups semantically related UI elements based on the XML screen hierarchy; (2) $\textbf{Co-planning}$, where local and cloud LLMs collaboratively identify the current sub-task; and (3) $\textbf{Co-decision-making}$, where the local LLM ranks relevant UI blocks, and the cloud LLM selects specific UI elements within the top-ranked block. CORE further introduces a multi-round accumulation mechanism to mitigate local misjudgment or limited context. Experiments across diverse mobile apps and tasks show that CORE reduces UI exposure by up to 55.6\% while maintaining task success rates slightly below cloud-only agents, effectively mitigating unnecessary privacy exposure to the cloud. The code is available at https://github.com/Entropy-Fighter/CORE.

Retrieval-Augmented Generation (RAG) significantly improves the performance of Large Language Models (LLMs) on knowledge-intensive tasks. However, varying response quality across LLMs under RAG necessitates intelligent routing mechanisms, which select the most suitable model for each query from multiple retrieval-augmented LLMs via a dedicated router model. We observe that external documents dynamically affect LLMs' ability to answer queries, while existing routing methods, which rely on static parametric knowledge representations, exhibit suboptimal performance in RAG scenarios. To address this, we formally define the new retrieval-augmented LLM routing problem, incorporating the influence of retrieved documents into the routing framework. We propose RAGRouter, a RAG-aware routing design, which leverages document embeddings and RAG capability embeddings with contrastive learning to capture knowledge representation shifts and enable informed routing decisions. Extensive experiments on diverse knowledge-intensive tasks and retrieval settings, covering open and closed-source LLMs, show that RAGRouter outperforms the best individual LLM and existing routing methods. With an extended score-threshold-based mechanism, it also achieves strong performance-efficiency trade-offs under low-latency constraints. The code and data are available at https://github.com/OwwO99/RAGRouter.

Online Internet platforms require sophisticated marketing strategies to optimize user retention and platform revenue — a classical resource allocation problem. Traditional solutions adopt a two-stage pipeline: machine learning (ML) for predicting individual treatment effects to marketing actions, followed by operations research (OR) optimization for decision-making. This paradigm presents two fundamental technical challenges. First, the prediction-decision misalignment: Conventional ML methods focus solely on prediction accuracy without considering downstream optimization objectives, leading to improved predictive metrics that fail to translate to better decisions. Second, the bias-variance dilemma: Observational data suffers from multiple biases (e.g., selection bias, position bias), while experimental data (e.g., randomized controlled trials), though unbiased, is typically scarce and costly --- resulting in high-variance estimates. We propose **Bi**-level **D**ecision-**F**ocused **C**ausal **L**earning (**Bi-DFCL**) that systematically addresses these challenges. First, we develop an unbiased estimator of OR decision quality using experimental data, which guides ML model training through surrogate loss functions that bridge discrete optimization gradients. Second, we establish a bi-level optimization framework that jointly leverages observational and experimental data, solved via implicit differentiation. This novel formulation enables our unbiased OR estimator to correct learning directions from biased observational data, achieving optimal bias-variance tradeoff. Extensive evaluations on public benchmarks, industrial marketing datasets, and large-scale online A/B tests demonstrate the effectiveness of Bi-DFCL, showing statistically significant improvements over state-of-the-art. Currently, Bi-DFCL has been deployed across several marketing scenarios at Meituan, one of the largest online food delivery platforms in the world.

This paper presents Nova, a real-time scheduling framework for serving agentic vision-language models (VLMs) on a single GPU with balanced per-request latency and overall request process throughput. Our design begins by enabling effective pipelining across vision encode, LLM prefill, and LLM decode stages of VLMs, by exploiting their heterogeneous resource demands during execution and incorporating elastic GPU spatial partitioning among stages to maximally utilize the compute and memory resources. Building on this, we introduce a realtime scheduling algorithm that adaptively calibrates resource allocation among stages based on a Pareto-optimal analysis of the latency-throughput trade-off, allowing the system to sustain responsiveness and resource efficiency under dynamic request loads. To further alleviate GPU memory pressure, we design a lightweight weight offloading strategy for vision encoders that preserves inference efficiency with minimized memory overhead. Extensive evaluations on both synthetic and real-world agent workloads demonstrate that Nova consistently outperforms the state-of-the-art baselines, improving the maximum latency by up to 23.3 %, while keeping competitive throughput.

Mobile video analysis systems often encounter various deploying environments, where environment shifts present greater demands for responsiveness in adaptations of deployed "expert DNN models". Existing model adaptation frameworks primarily operate in a cloud-centric way, exhibiting degraded performance during adaptation and delayed reactions to environment shifts. Instead, this paper proposes MOCHA, a novel framework optimizing the responsiveness of continuous model adaptation through hierarchical collaborations between mobile and cloud resources. Specifically, MOCHA (1) reduces adaptation response delays by performing on-device model reuse and fast fine-tuning before requesting cloud model retrieval and end-to-end retraining; (2) accelerates history expert model retrieval by organizing them into a structured taxonomy utilizing domain semantics analyzed by a cloud foundation model as indices; (3) enables efficient local model reuse by maintaining onboard expert model caches for frequent scenes, which proactively prefetch model weights from the cloud model database. Extensive evaluations with real-world videos on three DNN tasks show MOCHA improves the model accuracy during adaptation by up to 6.8% while saving the response delay and retraining time by up to 35.5× and 3.0× respectively.

The flourishing of Large Language Models (LLMs) calls for increasingly ultra-scale training. In this paper, we share our experience in designing, deploying, and operating our novel Astral datacenter infrastructure, along with operational lessons and evolutionary insights gained from its production use. Astral has three important innovations: (i) a same-rail interconnection network architecture on tier-2, which enables the scaling of LLM training. To physically deploy this high-density infrastructure, we introduce a distributed high-voltage direct current power system and a new air-liquid integrated cooling system. (ii) a full-stack monitoring system featuring cross-host and hierarchical logging correlation, which diagnoses failures at scale and precisely localizes root causes. (iii) an operator-granular forecasting component Seer that efficiently generates operator execution timelines with acceptable accuracy, aiding in fault diagnosis, model tuning, and network architecture upgrading. Astral infrastructure has been gradually deployed over 18 months, supporting LLM training and inference for multiple customers.

Root causing and failure localization are critical to maintain reliability in cloud network operations. When an incident is reported, network operators must review massive volumes of monitoring data and identify the root cause (i.e., error device) as fast as possible, making it extremely challenging even for experienced operators. Large language models (LLMs) have shown great potential in text understanding and reasoning. In this paper, we present BiAn, an LLM-based framework designed to assist operators in efficient incident investigation. BiAn processes monitoring data and generates error device rankings with detailed explanations. To date, BiAn has been deployed in our network infrastructure for 10 months and it has successfully assisted operators in identifying error devices more quickly, reducing time to root causing by 20.5% (55.2% for high-risk incidents). Extensive performance evaluations based on 17 months of real cases further demonstrate that BiAn achieves accurate and fast failure localization. It improves accuracy by 9.2% compared to the baseline approach.

Bayer-patterned color filter array (CFA) has been the go-to solution for color image sensors. In augmented reality (AR), although color interpolation (i.e., demosaicing) of pre-demosaic RAW images facilitates a user-friendly rendering, it creates no benefits in offloaded DNN analytics but increases the image channels by 3x inducing higher transmission overheads. The potential optimization in frame preprocessing of DNN offloading is yet to be investigated.

Exponentially expanding domain name system (DNS) over HTTPS (DoH) has significantly increased privacy but has also quietly masked malicious activities, rendering traditional threat detection systems meaningless. Existing deep learning-powered systems are unable to detect fleeting micro-temporal abnormalities in encrypted streams, are too costly for real-time operation, and are still vulnerable to adversarial attacks. To overcome these complex issues, this work proposes a new architecture—Neuromorphic Quantum Adversarial Learning (NQAL)—a bio-inspired, zero-knowledge-supported detection mechanism combining spiking neural networks (SNNs), quantum noise injection (QNI), and federated swarm intelligence to immunize, rather than detect, DoH-based attacks. The method relies on a neuromorphic model employing Dynamic Spiking Graph Attention (DSGAT) and Spike-Timing-Dependent Plasticity (STDP) to encode encrypted traffic as dynamic spike trains to enable ultra-fast, energy-efficient inference on processors such as Intel Loihi and BrainChip Akida. Quantum adversarial noise, emulated through stochastic perturbations created from quantum random walks, is injected during training to build gradient-obfuscating robustness. A threat immunization engine powered by adversarial GANs and quantum perturbations to simulate zero-day anomalies for preconditioning the model. Zero-knowledge verification is guaranteed through zk-SNARKs for privacy-preserving confirmation of anomalies without decrypting packets. Empirical studies confirm that NQAL achieves 99.18% accuracy, $$<1$$ ms latency, and 10x less energy consumption than GPU-based models, while also being robust to both classical and quantum adversarial attacks. Feasibility, novelty, and decentralization of the system amount to a paradigm shift from existing architectures—hence, making NQAL a resilient frontier in encrypted traffic immunization.

Monotonic graph algorithms, such as shortest path, BFS, and reachability, are fundamental to graph analytics and are widely used across domains. Recent systems employ pruning techniques to accelerate the processing of these algorithms. However, state-of-the-art monotonic graph engines are restricted to in-memory execution and cannot scale to graphs that exceed main memory capacity. In contrast, existing out-of-core graph engines are designed for general-purpose workloads and lack effective pruning mechanisms tailored to monotonic graph algorithms. To bridge this gap, we present Gem, an out-of-core graph engine designed for monotonic graph algorithms. Gem introduces a PageRank-based graph sketch that captures key topological features inmemory with minimal preprocessing overhead. Building on this sketch, we propose a novel graph abstraction that enables the direct derivation of tight bounds for monotonic graph algorithms, supporting effective pruning at both the vertex and partition levels. Comprehensive evaluations on six real-world datasets, including the 42.5-billion-edge ClueWeb graph, show that Gem significantly outperforms existing systems. It achieves up to 135.40× speedup over GridGraph and 12.58× over Wonderland in out-of-core settings, and also delivers substantial improvements in other modes: up to 10.41× over RisGraph in memory and 20.64× over CGgraph out-of-GPU memory.

Learned indexes, which model key-value data structures by machine learning models, have been extensively studied. However, the fastest immutable learned indexes (e.g., RMI) do not provide the same tight lookup bounds as classical indexes such as B-trees. There are learned indexes that provide tight bounds (e.g., PGM) but those fall short in query performance. This gives rise to an interesting open question: whether there exists a learned index that simultaneously achieves state-of-the-art empirical performance and matching complexity? In this paper, we give a positive answer to this standing problem.We propose two new online model-building policies: (1) simplifying distribution by the adoption of a proper granularity (i.e., grouping multiple keys together for model-building) and (2) actively tuning distribution through key repositioning. Additionally, we introduce a general framework that combines these two policies for performance optimization under a given memory budget. We put everything together to design VEGA, a learned index that simultaneously achieves competitive theoretical and empirical performance compared to state-of-the-art learned indexes. We conducted extensive evaluations, demonstrating VEGA achieves both better lookup and building performance.

ABSTRACT Context The rise of artificial intelligence of things (AloT) has enabled smart cities and industries, and UAV‐assisted edge computing networks are an important technology to support the above scenarios. UAV‐assisted refers to leveraging UAVs as a dynamic, flexible infrastructure to assist edge network data processing and communication tasks. Multiple UAVs can use their own resources, and collaborate edge servers to train artificial intelligence (Al) models. Objective Compared with cloud‐based collaborative computing scenarios, UAV‐assisted edge collaborative learning can reduce training and inference delays and improve user satisfaction. However, UAV‐assisted edge networks scenario brings new challenges in terms of transmission burden and energy consumption. Method This paper proposes a prototype‐based joint optimization and training software system. The system consists of an optimization module and a training module. The optimization module first models an optimization problem including energy consumption and prototype error. Then it solves the optimization problem by problem transformation and plans the location of each UAV given the objects' position. After UAVs fly to the designated area and complete data collection, UAVs and the edge server train a model according to the proposed prototype‐based collaborative training module. Our training module enables multiple UAVs and an edge server to collaboratively train a model by lightweight prototype transmission and prototype aggregation. We also prove the convergence of the proposed collaborative training method. Results Results show our method reduces prototype error and energy consumption by at least 12.31% and improves model accuracy by 3.62% with a little communication burden. Conclusion Finally, we verify system performance through experiments.

With the improvement of edge-based autonomous systems such as mobile Industrial IoT (IIoT) networks, edge devices can capture and upload videos with increasing bitrates. Massive edge-computing end nodes are eager for adequate multimedia data to satisfy the requirements of real-time video services. However, existing encoding standards for video services in Web 2.0 are specifically designed for something other than IoT video streaming. We have improved our Adaptive Compression-Reconstruction (ACORN) framework to obtain ACORN+, based on compressed sensing and recent advances in deep learning. At end nodes, we compress multiple sequential video frames into a single frame to reduce video volume. Given that multiple kinds of intelligent tasks are expected to be finished on the device side, we also designed a device-cloud collaboration scheme where deep learning-based algorithms can be executed on both the device and server sides. Experiments reveal that video analytics can be conducted on compressed frames. Taking action recognition as a device-cloud collaboration use case, we find ACORN \(+\) obtains more than 3 \(\times\) speedup on compressed frames. The reconstruction algorithm in ACORN \(+\) is with 1– 4 dB improvements. Moreover, the encoding time cost and the encoded video volume are reduced by more than 4 \(\times\) under the ACORN \(+\) framework. 1

3D Gaussian splatting (3DGS) suggests the use of explicit point-based 3D representations for high-fidelity novel view synthesis, with training and rendering speeds that are better than prior neural radiance fields. However, 3DGS relies heavily on synthetic point clouds generated by structure from motion (SfM) or multi view stereo (MVS) techniques, and lacks a well-defined method to initially condition them. In this work, we first propose a mesh-aligned method that attaches the 3D Gaussian to the extracted surface meshes for reliable initialization during training, and then directly manipulates the learnable 3DGS in the local coordinate using the triangular meshes. In addition, to constrain the stereo Gaussian on a planar mesh for better rendering, both normal and depth losses are designed to optimize the orientation and significance of the Gaussian. In practice, we further apply this method to multi-resolution scene rendering while resolving the aliasing effect. Unlike directly changing the size and number of Gaussians that may interfere the rendering quality at different resolutions, we argue that the properties of the modelled mesh are naturally resistant to aliasing effects. By utilising triangular meshes for Gaussian binding and adaptive learning, the proposed method can maintain high-fidelity rendering after splatting on multi-resolution concrete images. Extensive experiments demonstrate the effectiveness of our approach and its advantages over single full-resolution baseline.

Estimating per-flow cardinality from high-speed data streams has many applications such as anomaly detection and resource allocation. Yet despite tracking single flow cardinality with approximation algorithms offered, there remain algorithmical challenges for monitoring multi-flows especially under unbalanced cardinality distribution: existing methods adopt a uniform sketch layout and incur a large memory footprint to achieve high accuracy. Furthermore, they are hard to implement in the compact hardware used for line-rate processing. In this paper, we propose Couper, a memory-efficient measurement framework that can estimate cardinality for multi-flows under unbalanced cardinality distribution. We propose a two-layer structure based on a classic coupon collector's principle, where numerous mice flows are confined to the first layer and only the potential elephant flows are allowed to enter the second layer. Our two-layer structure can better fit the unbalanced cardinality distribution in practice and achieve much higher memory efficiency. We implement Couper in both software and hardware. Extensive evaluation under real-world and synthetic data traces show more than 20× improvements in terms of memory-efficiency compared to state-of-the-art.

The map matching of cellular data reconstructs real trajectories of users by exploiting the sequential connections between mobile devices and cell towers. The difficulty in obtaining paired cellular-GPS data and the cellular variation compromise the accuracy and reliability of existing map matching approaches. In this paper, we propose a novelunsupervisedconvolutional network for cellularmapmatching (UCMM) to address these challenges. UCMM employs a dual encoder-decoder network to capture a shared representation from both the cellular and GPS domains in an unsupervised manner. It leverages a dedicated convolutional architecture to tackle the varying lengths of output sequential data. An attention mechanism is specially introduced to deal with the cellular variation. The effectiveness of UCMM is demonstrated through comprehensive evaluations, which show that UCMM achieves a substantial improvement in matching accuracy and deduction of training time compared with the best-known prior works. These improvements make UCMM a significant advancement in the field of map matching.

Augmented reality (AR) offers users immersive experiences to interact with digital contents in their physical space. However, practical AR applications are challenged by the tight coupling of algorithm and engineering during the development and deployment phases as well as the execution requirements of hybrid AR subtasks on heterogeneous and resource-constraint mobile devices. In this work, we build an end-to-end, cross-platform, and efficient AR system, called ARSys. The infrastructure in ARSys adopts the new principle of integrated design, unifies and refines AR fundamental capabilities, supports streaming media processing, model inference, and real-time rendering by exposing high-performance tensor compute engine to top, and constructs a Python multi-instance virtual machine as the cross-platform AR task execution container. The runtime mechanism of ARSys schedules AR tasks in a pipeline parallelism way and allocates subtasks to hardware backends by optimizing the slowest node. The development workbench and the deployment platform in ARSys allow the decoupling of algorithms written in Python from engineering components in C/C++ and further support remote debugging and quick validation of AR algorithms. We extensively evaluate ARSys in practical AR applications across high-end, mid-end, and low-end Android and iOS devices, demonstrating higher development, deployment, and runtime efficiency than existing MediaPipe-oriented framework. ARSys has been integrated into Mobile Taobao for production use.

Large language models (LLMs) exhibit remarkable capabilities in natural language understanding and generation. However, the accuracy of the inference depends deeply on the contexts of queries, especially for personal services. Abundant mobile sensing data collected by sensors embedded in smart devices can proactively capture real-time user contexts. However, raw sensing data are low-quality (e.g., existing missing data and data redundancy) and are incapable of providing accurate contexts. In this work, we present ConGen, a user context generation framework for LLMs, aiming at prompting users’ contexts through their implicit mobile sensing information. ConGen integrates two components: refined data completion and multi-granularity context compression. Specifically, the refined data completion couples data-centric feature selection by leveraging the eXplainable AI (XAI) method into the data imputation model to generate fewer but more informative features for efficient and effective context generation. Additionally, we implement multi-granularity context compression, reducing timestep- and context-level data redundancy while further elevating context quality. Experiment results show that ConGen can generate more accurate context, surpassing competitive baselines by 1.3%-8.3% in context inference on all four datasets. Moreover, context compression significantly reduces redundancy to$1/70\sim 1/40$of the original data amount, and further improves the context accuracy. Finally, the enhanced performance of LLMs, as demonstrated by both quantitative and qualitative evaluations of prompting ConGen-generated user contexts, underscores the effectiveness of ConGen.

In many real-life applications, wireless chargers are deployed outdoor or in public area or even unattended environment such as hotels, restaurants, retail stores. They are exposed to various risks and malicious attacks that may break them down and further incur significant cost (e.g., battery replacement and maintenance) or performance degradation. Hence, we consider the problem ofFault-tolerant wIreless chaRger placeMent (FIRM): given a set of wireless chargers and a set of tasks to be collaboratively conducted by a set of rechargeable devices, determining where to deploy the chargers to maximize the worst-cast overall task charging utility subject to the constraint that up to$\tau$chargers may break down. FIRM is a non-linear combinatorial two-level optimization problem. We first consider a relaxed version of FIRM (FIRM-R for short) corresponding to the inner optimization problem in FIRM. To address FIRM-R, we first propose an area discretization scheme to convert the infinite solution space into finite candidate positions. We then devise a power allocation method, based on which we prove that FIRM-R falls into the realm of maximizing a monotone submodular function under a uniform constraint. We then propose a constant-factor approximation algorithm to solve FIRM-R. Taking the above approximation algorithm as a subroutine, we further develop an approximation algorithm that solves FIRM with a constant-factor approximation ratio. Our extensive simulations and field experiments demonstrate that the overall charging utility of our proposed algorithm FIRM considering fault tolerance by greedy removal of$\tau$chargers outperforms the that of FIRM-R without considering fault tolerance by greedy removal of$\tau$chargers by at least 119.89%.

For Unmanned Aerial Vehicles (UAVs) monitoring tasks, capturing high quality images of target objects is important for subsequent recognition. Concerning the problem, many prior works study placement/trajectory planning for UAVs to maximize the quality of captured images. However, all of them overlook a fact thatUAV monitoring may cause a huge risk/annoyance on living objects.In this paper, we investigate the novel problem of oPtimizing uncrewed aErial vehicles plAcement byConsidering both monitoring utility and adverseEffects (PEACE). We propose an approach to solve PEACE, which is proved to be NP-hard. Overall, our approach achieves a$1- \frac{1}{e}-\varepsilon$approximation ratio. First, we approximate the original problem of PEACE as a classical problem of Monotone Submodular function Maximization under a Uniform Matroid constraint (MSMUM) with a controlled gap. Then, for MSMUM, we propose a combination of algorithms achieving a$1-\frac{1}{e}$approximation and$O(n\log n)$time complexity considering the correlation among the UAV monitoring strategies. The proposed algorithms outperform existing algorithms for MSMUM through theoretical analysis and experimental results. Extensive simulations and field experiments demonstrate the effectiveness of our approach, achieving performance gains of 9.0% to 1434.5% compared to existing methods.

With the popularization of smart phones, mobile crowdsourcing emerged and gained growing attention in the recent years. Mobile users are now able to conduct complex tasks with the communication between each other. In this paper, we study the task scheduling problem in the mobile crowdsourcing systems based on the spontaneously formed mobile social networks (MSNs). We introduce two crowdsourcing task scheduling problems under this system model, with one problem aiming to minimize the total cost of some crowdsourcing tasks and the other focusing on minimizing the final completion time of the tasks belonging to the same project. We introduce a unified framework to solve the problems and propose two approximation algorithms for these two problems in the offline versions respectively and prove Their approximation ratios accordingly. Based on the two algorithms, we further design two online algorithms to deal with the workers' dynamism and also analyze their competitive ratios. Finally, we verify the effectiveness and efficiency of the proposed methods through numerical experiments on real and synthetic datasets.

Federated learning (FL) can be implemented in large-scale wireless networks in a hierarchical way, introducing edge servers as relays between the cloud server and devices. These devices are dispersed within multiple clusters coordinated by edges. However, the devices are typically mobile users with unpredictable trajectories, and the impact of their mobility on the model training process is not well-studied. In this work, we propose a newMobIlity-Driven feDeratedLEarning framework, namely MIDDLE. MIDDLE addresses unbalanced model updates by capitalizing on model aggregation opportunities on mobile devices due to their mobility across edges. It consists of two components: on-device model aggregation, which aggregates models from different edges carried by mobile devices as they move across edges, and in-edge device selection, adjusting the current edge optimization direction through careful device selection. Theoretical analysis emphasizes that on-device model aggregation can reduce bias in model updating on edges and the cloud, thereby accelerating the FL model convergence. Building on this analysis, we introduce on-device global control averaging, modifying the training process on mobile devices and extending MIDDLE into$\text{MIDDLE}^{+}$. Extensive experimental results validate that MIDDLE and$\text{MIDDLE}^{+}$can reduce the time steps to reach the target accuracy by 19.44% and 20.37% at least, respectively.

Traditional fine-grained classification focuses on visible light domains, such as animals and cars. However, these methods often perform poorly when applied to radar images and images of satellites because of challenges such as distinguishing between noise and objects and the significant scale differences among object components. To address these unique scenarios, we propose the scale-shift attention in polarization domain (SAPD) method for fine-grained classification in satellite ISAR images. Specifically, radar emits different types of waves, each with distinct imaging effects. We utilize multipolarization inputs and introduce a polarization domain query module to integrate complementary features from various radar wave types captured from the same viewpoint. This multipolarization learning helps distinguish noise and leverages complementary features from different inputs. Moreover, to handle the substantial scale differences between centimeter-level payloads and the overall meter-level structure of satellites, we propose a scale-shift attention mechanism based on shift kernels. This mechanism extends attention in the direction specified by the shift kernel by incorporating adjacent pixels, allowing for the diffusion of attention. This is beneficial for capturing features of satellite components with varying scales and shapes. Extensive experiments on a novel satellite ISAR image dataset validate the effectiveness and superiority of the SAPD.

With the increase of diversity in application preferences and networks, existing congestion control algorithms (CCAs) do not accommodate this complicated reality. Previous classic CCAs are designed for a specific domain with fixed rules, failing to adapt to such diversities. Recently surged learning-based CCAs have great potential in adaptability and flexibility but are not practical due to unsatisfying performance on convergence, fairness, overhead, consistency and safety assurance. In this paper, we propose Libra, a unified congestion control framework, that can empower these properties by combining the wisdom of classic and reinforcement learning (RL)-based CCAs. Extensive evaluation of Libra’s Linux kernel implementations on both live Internet and emulated networks shows performance improvement under dynamic networks (e.g.,$1.2\times $throughput than Orca on average). At the same time, Libra can flexibly satisfy different application needs, reduce the running overhead by at most$0.88\times $and perform good fairness and convergence properties, well-fitting our theoretical analysis.

While the flexibility of programmable switches brings opportunities, it also introduces security risks. Hence, it is vital to conduct effective troubleshooting in the programmable switch to mitigate frequent network failures. However, troubleshooting programmable switch failures is challenging due to their enhanced flexibility and functionality compared to regular switches, posing increased difficulty in debugging, particularly with limited debugging tools and information. To address this problem, we propose an efficient troubleshooting method that records real-time information about packets in the data plane, including the tables involved in packet processing. Unfortunately, due to hardware limitations, it is infeasible to record all tables’ information in the data plane. Thus, the key is to find the table set reflecting the execution path a packet goes through while minimizing the resource overhead. We first represent P4 programs as a probabilistic transition directed acyclic graph (DAG) and employ information entropy to quantify the information within a set of tracked tables. Then, we adopt a two-step approach and design algorithms to find both optimal and approximately optimal table record plans. The evaluation results show the efficacy of the proposed method, including achieving the same path recovery rate as the related works with less than one-third of the resource consumption.

Congestion control protocols play a vital role in enhancing the performance of various applications within datacenter networks. While reactive congestion control (RCC) protocols are widely deployed in commercial datacenters, the research community has actively explored token-based proactive congestion control (TCC) protocols to further push the boundaries of performance. However, despite the emergence of numerous TCC variants, there has been a lack of systematic exploration in the design space of TCC. This paper aims to bridge this gap by proposing a framework for understanding the design choices within the TCC approach. In this study, we systematically analyze different design choices of TCC approaches and leverage this understanding to develop a novel TCC protocol called ToCC. To implement ToCC, we address a set of challenges and deploy it in NP-based smart NICs. We compare ToCC with state-of-the-art TCC and RCC protocols through extensive large-scale simulations and testbed evaluations. The results demonstrate that ToCC exhibits robustness in achieving low latency across various scenarios. Additionally, ToCC effectively reduces buffer occupancy by 4.8 times compared to existing approaches, and under incast scenarios, it significantly shortens flow completion time by up to 90%. Congestion control protocols are crucial for optimizing the performance of datacenter network applications. Although reactive congestion control (RCC) protocols are commonly used in commercial datacenters, researchers have been exploring token-based proactive congestion control (TCC) protocols to further enhance network performance. Despite the development of numerous TCC variants, there has not been a thorough examination of the design space of TCC protocols until now. This paper aims to address this gap by introducing a framework for understanding the design choices within the TCC approach for TCC protocols. By analyzing various design aspects of TCC approaches, we create a novel TCC protocol called ToCC. At the central of ToCC design is that it leverages congestion control mechanisms over tokens. To implement ToCC, we tackle several challenges and integrate it into NP-based smart NICs. Comparing ToCC with state-of-the-art TCC and RCC protocols through extensive large-scale simulations and testbed evaluations, we find that ToCC consistently achieves low latency across different scenarios. Moreover, ToCC significantly reduces buffer occupancy by 4.8 times compared to existing methods, and during incast scenarios, it decreases flow completion time by up to 90%.

Approximate membership query (AMQ) data structures can approximately determine whether an element exists in a given dataset. They are widely used in parallel and distributed systems (e.g., high-performance databases, distributed cache systems, and bioinformatics systems) to avoid unnecessary dataset accesses, thereby accelerating massive data processing. For AMQ data structures used in the above systems, achieving high throughput, low false positive rate, and large capacity objectives simultaneously is critical but challenging. Porting AMQ data structures from DRAM to persistent memory makes it possible to achieve the above three objectives simultaneously, but this porting is not a trivial task. Specifically, existing AMQ data structures generate numerous random accesses and/or sequential writes on persistent memory, resulting in poor throughput. Therefore, in the conference version of this paper, we proposed a novel AMQ data structure called wormhole filter, which achieves high throughput on persistent memory, thereby achieving the above three objectives simultaneously. In this journal version, we extend our prior work by introducing parallel wormhole filters to enhance parallel performance. Additionally, we integrate parallel wormhole filters into the LevelDB database system to show that porting AMQ data structures to persistent memory significantly improves system endto-end throughput. Theoretical analysis and experimental results show that wormhole filters significantly outperform state-of-theart AMQ data structures. For example, wormhole filters achieve 12.06× insertion throughput, 1.98× positive lookup throughput, and 8.82× deletion throughput of the best competing baseline.

As one of the pillars in cluster computing frameworks, coflow scheduling algorithms can effectively shorten the network transmission time of cluster computing jobs, thus reducing the job completion times and improving the execution performance. However, most of existing coflow scheduling algorithms failed to consider the influences of concurrent flows, which can degrade their performance under a massive number of concurrent flows. To fill the gap, we propose a unified communication agent named Courier to minimize the number of concurrent flows in cluster computing applications, which is compatible with the mainstream coflow scheduling approaches. To maintain the scheduling order given by the scheduling algorithms, Courier merges multiple flows between each pair of hosts into a unified flow, and determines its order based on that of origin flows. In addition, in order to adapt to various types of topologies, Courier introduces a control mechanism to adjust the number of flows while maintaining the scheduling order. Extensive large-scale trace-driven simulations have shown that Courier is compatible with existing scheduling algorithms, and outperforms the state-of-the-art approaches by about 30% under a variety of workloads and topologies.

Discrete-event simulation (DES) is a prevalent tool for evaluating network designs. Although DES offers full fidelity and generality, its slow performance limits its application. To speed up DES, many network simulators employ parallel discrete-event simulation (PDES). However, adapting existing network simulation models to PDES requires complex reconfigurations and often yields limited performance improvement. In this paper, we address this gap by proposing a parallel-efficient and user-transparent network simulation kernel, Unison, that adopts fine-grained partition and load-adaptive scheduling optimized for network scenarios. We prototype Unison based on ns-3. Existing network simulation models of ns-3 can be seamlessly transitioned to Unison. Testbed experiments on commodity servers demonstrate that Unison can achieve a 40× speedup over DES using 24 CPU cores, and a 10× speedup compared with existing PDES algorithms under the same CPU cores.

Approximate membership query (AMQ) data structures can approximately determine whether an element is in the set with high efficiency. They are widely used in distributed systems, database systems, bioinformatics, IoT applications, data stream mining, etc. However, the memory consumption of AMQ data structures grows rapidly as the data scale grows, which limits the system's ability to process a massive amount of data. The emerging persistent memory provides a close-to-DRAM access speed and terabyte-level capacity, facilitating AMQ data structures to handle massive data. Nevertheless, existing AMQ data structures perform poorly on persistent memory due to intensive random accesses and/or sequential writes. Therefore, we propose a novel AMQ data structure called wormhole filter, which achieves high performance on persistent memory by reducing random accesses and sequential writes. In addition, we reduce the number of log records for lower recovery overhead. Theoretical analysis and experimental results show that wormhole filters significantly outperform competitive state-of-the-art AMQ data structures. For example, wormhole filters achieve 23.26× insertion throughput, 1.98× positive lookup throughput, and 8.82× deletion throughput of the best competing baseline.

Hierarchical Federated Learning (HFL) is a practical implementation of federated learning in mobile edge computing, employing edge servers as intermediaries between mobile devices and the cloud server for device coordination and cloud communication. However, the devices are usually mobile users with unpredictable mobile trajectories and statistical heterogeneity, leading to the edge models optimized along dynamic edge data distribution directions and further resulting in instability and slow convergence of the global model. In this work, we propose a Mobility-Aware deviCe sampling algorithm in HFL, namely MACH, which can dynamically maintain the device sampling strategy at each edge to accelerate the convergence of the global model. First, we analyze the convergence bound of HFL with mobile devices under arbitrary device sampling probabilities. Based on this convergence bound, we formalize the sampling optimization problem for mobility-aware device sampling, aiming to minimize the convergence error under time-averaged cost constraints, while taking the limited device-edge wireless channel capacity into account. Next, we introduce the MACH algorithm, consisting of two underlying components: experience updating and edge sampling. Experience updating utilizes an upper confidence bound method to estimate device statistical information online, and edge sampling customizes a sampling strategy on each edge based on the estimated device statistical information. Finally, extensive experimental results through real-world mobile device trajectories validate that MACH can reduce the time required to achieve a target accuracy by 25.00% - 56.86%.

The shipping industry has irreplaceable importance in international trade and commerce. How to dynamically price different containers has long been a hot topic due to its direct connection to the final revenue. Two critical observations have been made after a comprehensive survey within a top liner company, China Ocean Shipping Company (COSCO). (1) Each type of container carried on a liner ship has its maximum capacity. (2) The sales volume is occasionally subject to huge fluctuations due to rare uncontrollable factors, such as COVID. Based on the above two points and the liner routine's periodic nature, we model the dynamic pricing problem as an episodic MDP model integrating with both capacity constraints and adversarial corruption, named C3-MDP. To maximize the cumulative revenue in the C3-MDP setting, we propose a programming framework, Bonus-Exploration based Episodic Programming (BEEP). This framework can directly accommodate the linear programming algorithm to form the algorithm BEEP-LP, which provides the episode-wise greedy optimal strategy. Furthermore, a detailed regret analysis is provided, showing that BEEP-LP has a regret that is sublinear in the number of episodes. Combining deep techniques, we also present an approximation algorithm BEEP-DQN in the case of large state-action space to strike a balance between the running time and the performance. Abundant experiments based on real container sales data exhibit the rationality of C3-MDP and the effectiveness of BEEP.

Sketches have gained popularity as effective methods for estimating frequency in data streams, and optimizing their accuracy is critical in many applications. However, while sketches are backed by a standard guarantee under the worst-case analysis, their actual errors can vary significantly with real-world skewed data streams. Therefore, it is challenging to configure sketches to optimize accuracy without prior knowledge of the input. Moreover, even with a new configuration, it is unclear when to apply it. This paper presents a novel sketch framework that can be dy-namically configured to optimize the accuracy given a processed data stream. Specifically, we provide a precise guarantee and derive an optimal number of hash functions under the Zipfian distribution, which is an appropriate way to model skewed data streams in practice. We then propose a dynamically configurable sketch framework, namely DISCO, that can estimate the distri-bution parameter and adjust the number of hash functions on the fly to optimize accuracy. We provide rigorous mathematical analysis and apply DISCO to three classical solutions, including the Count-min, Conservative Update, and Count sketches. Experimental results, using synthetic and real datasets, show that DISCO can achieve the optimal configuration for the metric (i.e., FP) related to the sketch guarantee, while achieving near-optimal accuracy for other common metrics (e.g., ARE) compared with state-of-the-art methods.

Traffic flow prediction is a critical issue in transportation engineering and presents distinct challenges when handling large-scale datasets in the real world. Existing complex spatio-temporal forecasting paradigms use the same parameters to fit traffic sequences with varying spatio-temporal features, and tend to train an average performance model over different time series. This approach greatly reduces their accuracy when applied to larger road networks. Moreover, the significant differences in traffic data distribution from one city to another can also pose great challenges. The same model may be excellent for one city and mediocre when applied to another. To this end, we propose a Meta Backbone Agnostic Spatio-Temporal Clustering Framework for Traffic Forecasting on Large-Scale Road Networks named MetaSTC. We tackle the disparities of spatio-temporal features of traffic flow through a spatio-temporal clustering-based strategy. We design meta-learner for large-scale road network that dynamically extracts the shared information across roads in the same sub-task. In this way, the model can represent task-specific details with a simpler model and make quick and accurate predictions. Our paradigm is backbone-agnostic and can be combined with different traffic prediction models, solving the problem caused by the difference in data distribution. Extensive experimental results conducted on real-world traffic dataset demonstrate the high accuracy and computational efficiency of our model over SOTA approaches.

To bring the great power of modern DNNs into mobile computing and distributed systems, current practices primarily employ one of the two learning paradigms: cloud-based learning or on-device learning. Despite their distinct advantages, neither of these two paradigms could effectively deal with highly dynamic edge environments reflected in quick data distribution shifts and on-device resource fluctuations. In this work, we propose Nebula, an edge-cloud collaborative learning framework to enable rapid model adaptation for changing edge environments. To achieve this, we first propose a new block-level model decomposition scheme to decompose the large cloud model into multiple combinable modules. With this design, we can agilely derive personalized sub-models with compact sizes for edge devices, and quickly aggregate the updated sub-models to integrate new knowledge learned on the edge into the cloud model. We further propose an end-to-end learning framework that incorporates the modular model design into an efficient model adaptation pipeline, including an offline on-cloud model prototyping and training stage, and an online edge-cloud collaborative adaptation stage. Extensive experiments demonstrate that Nebula improves model performance (e.g., 18.89% accuracy increase) and resource efficiency (e.g., 7.12 × communication cost reduction) in adapting models to dynamic edge environments.

The sweep coverage problem is one of the important issues in spatial crowdsourcing, which requires task participants to monitor a series of Points of Interest (PoIs) periodically. In this paper, we study the Cooperative Sweep Coverage (CSC) problem with the objective of minimizing the maximum sweep period. We propose an iterative clustering algorithm based on spherical projection, called SP-Cycle. The algorithm firstly projects the points in the 2D space to a spherical surface in the 3D space. It then utilizes a new balancing clustering algorithm and uses an iterative coordinate updating method in the 3D spherical space based on the gradient descent, the aim of which is to use the simple minimum spanning tree length computation in the spherical surface to replace the complex Traveling Salesman Problem (TSP) cycle computation in the original 2D space, which improves the performance while keeping a low computational complexity. After we get the clusters, we can compute the lengths of TSP cycles and enter a new iteration. Experimental results based on synthetic and real-world datasets demonstrate the effectiveness of our proposed algorithm. The code is available at https://github.com/GaoYucen/CSC.

In this paper, we consider the problem of approximate membership testing problem on skewed network traffic traces, in which some hot or popular items repeat frequently. Previous solutions suffer from either high false positive rates or low lookup throughput. To address this problem, we propose a variant of the cuckoo filter, enhanced with a hotness-aware suffix cache. We note that a false positive item must have a matched fingerprint in the cuckoo filter, and propose to reduce false positives by memorizing them, but with their suffixes only. For each false positive item, we apply a linear-congruential-based hash function and then divide the hash value into three parts: the bucket index to be accessed in the cuckoo filter, the fingerprint to be stored in the cuckoo filter, and the suffix to be cached. Combining the three parts, we propose RCF that can uniquely identify a hot false positive item, which thus reduces hot false positives. Our evaluation results indicate that RCF significantly outperforms non-adaptive filters on skewed data traces. Given the same memory size, it achieves a much lower false positive ratio without sacrificing its lookup throughput. Compared with adaptive filters, RCF provides a competitive false positive ratio while offering a considerably higher lookup throughput.

Many billion-scale large language models (LLMs) have been released for resource-constraint mobile devices to provide local LLM inference service when cloud-based powerful LLMs are not available. However, the capabilities of current on-device LLMs still lag behind those of cloud-based LLMs, and how to effectively and efficiently enhance on-device LLM inference becomes a practical requirement. We thus propose to collect the user's historical interactions with the cloud-based LLM and build an external datastore on the mobile device for enhancement using nearest neighbors search. Nevertheless, the full datastore improves the quality of token generation at the unacceptable expense of much slower generation speed. To balance performance and efficiency, we propose to select an optimal subset of the full datastore within the given size limit, the optimization objective of which is proven to be submodular. We further design an offline algorithm, which selects the subset after the construction of the full datastore, as well as an online algorithm, which performs selection over the stream and can be flexibly scheduled. We theoretically analyze the performance guarantee and the time complexity of the offline and the online designs to demonstrate effectiveness and scalability. We finally take three ChatGPT related dialogue datasets and four different on-device LLMs for evaluation. Evaluation results show that the proposed designs significantly enhance LLM performance in terms of perplexity while maintaining fast token generation speed. Practical overhead testing on the smartphone reveal the efficiency of on-device datastore subset selection from memory usage and computation overhead.

Complex event recognition (CER) refers to identifying specific patterns composed of several primitive events in event stores. Since full-scanning event stores to identify primitive events holding query constraint conditions will incur costly I/O overhead, a mainstream and practical approach is using index techniques to obtain these events. However, prior index-based approaches suffer from significant I/O and sorting overhead when dealing with high predicate selectivity or long query window (common in real-world applications), which leads to high query latency. To address this issue, we propose ACER, a Range Bitmap-based index, to accelerate CER. Firstly, ACER achieves a low index space overhead by grouping the events with the same type into a cluster and compressing the cluster data, alleviating the I/O overhead of reading indexes. Secondly, ACER builds Range Bitmaps in batch (block) for queried attributes and ensures that the events of each cluster in the index block are chronologically ordered. Then, ACER can always obtain ordered query results for a specific event type through merge operations, avoiding sorting overhead. Most importantly, ACER avoids unnecessary disk access in indexes and events via two-phase filtering based on the window condition, thus alleviating the I/O overhead further. Our experiments on six real-world and synthetic datasets demonstrate that ACER reduces the query latency by up to one order of magnitude compared with SOTA techniques.

Microservice architecture has become a driving force in enhancing the modularity and scalability of web applications, as evidenced by the Alipay platform's operational success. However, a prevalent issue within such infrastructures is the suboptimal utilization of CPU resources due to inflexible resource allocation policies. This inefficiency necessitates the development of dynamic, accurate workload prediction methods to improve resource allocation. In response to this challenge, we present STAMP, a Spatio Temporal Graph Network for Microservice Workload Prediction. STAMP is designed to comprehensively address the multifaceted interdependencies between microservices, the temporal variability of workloads, and the critical role of system state in resource utilization. Through a graph-based representation, STAMP effectively maps the intricate network of microservice interactions. It employs time series analysis to capture the dynamic nature of workload changes and integrates system state insights to enhance prediction accuracy. Our empirical analysis, using three distinct real-world datasets, establishes that STAMP exceeds baselines by achieving an average boost of 5.72% in prediction precision, as measured by RMSE. Upon deployment in Alipay's microservice environment, STAMP achieves a 33.10% reduction in resource consumption, significantly outperforming existing online methods. This research solidifies STAMP as a validated framework, offering meaningful contributions to the field of resource management in microservice architecture-based applications.

To provide full-body 3D videos of performers showcasing diverse clothes with dynamic movements for e-commerce platforms, we develop an end-to-end, low-cost, and high-fidelity production and deployment pipeline. We first set up a low-cost capture studio with only 24 RGB cameras and embrace fast neural surface reconstruction to produce high-quality meshes without depth information. We then quickly group all the frames with local motion priors, select a keyframe for each group, and accurately register any other frame to the keyframe under the guidance of semantic labels, thereby avoiding transmitting all the frames to mobile devices and loading them into memory. For real-time rendering, we propose an on-device sparse computation method for efficient deformation from keyframes to the other frames. Evaluation over 2 self-captured performances and 8 public performances reveals that the pipeline achieves the reconstruction time of 28 minutes per frame, the average PSNR of 30.4, the average bandwidth requirement of 4.2MB/s, and the on-device frame rate of 60 fps, demonstrating superiority over existing baselines.

In modern mobile applications, users frequently encounter various new contexts, necessitating on-device continual learning (CL) to ensure consistent model performance. While existing research predominantly focused on developing lightweight CL frameworks, we identify that data scarcity is a critical bottleneck for on-device CL. In this work, we explore the potential of leveraging abundant cloud-side data to enrich scarce on-device data, and propose a private, efficient and effective data enrichment framework Delta. Specifically, Delta first introduces a directory dataset to decompose the data enrichment problem into device-side and cloud-side sub-problems without sharing sensitive data. Next, Delta proposes a soft data matching strategy to effectively solve the device-side sub-problem with sparse user data, and an optimal data sampling scheme for cloud server to retrieve the most suitable dataset for enrichment with low computational complexity. Further, Delta refines the data sampling scheme by jointly considering the impact of enriched data on both new and past contexts, mitigating the catastrophic forgetting issue from a new aspect. Comprehensive experiments across four typical mobile computing tasks with varied data modalities demonstrate that Delta could enhance the overall model accuracy by an average of 15.1%, 12.4%, 1.1% and 5.6% for visual, IMU, audio and textual tasks compared with few-shot CL, and consistently reduce the communication costs by over 90% compared to federated CL.

Two-stage bandit-based algorithms have found widespread application in modern online platforms, offering a balance between cost and accuracy. The initial stage involves coarse filtering of a small candidate set of promising items from a large corpus, while the subsequent stage refines the selection and presents a single item to the user. In this work, to the best of our knowledge, we for the first time undertake a theoretical analysis of the two-stage stochastic multi-armed bandit problem. Specifically, we model the two-stage bandit problem as a two-stage online optimization, and conduct a theoretical analysis. We demonstrate that while the optimization objective of the first stage may seem intuitive, it is, in fact, non-trivial. We devise a proxy optimization objective, emphasize the importance of a carefully designed exploration strategy, and establish the theoretical analysis for the application of Upper Confidence Bound (UCB)-based algorithms in the first stage. Furthermore, we provide a regret analysis of the proposed two-stage bandit algorithm, demonstrating a gap-dependent upper bound of [EQUATION], where [EQUATION] is the largest reward gap, and a gap-independent lower bound of [EQUATION], where n represents the horizon.

The Internet of Things (IoT) has expanded its reach through applications like environmental monitoring and smart agriculture. Its growth is often limited by the need for extensive IoT gateways and base stations. Satellite-based IoT, using low Earth orbit (LEO) satellites, offers a solution by enabling global sensing without terrestrial infrastructure. This technology allows IoT devices to transmit data to satellite gateways, which is crucial for dynamic environments like the ocean where nodes drift with currents. In such applications, position information is valuable for data analysis and tracing. Traditional GPS modules, while affordable and low-energy, can be inefficient due to the large payload space required for precise positioning data. The payload used to report the position might be as long as the sensed data itself. This paper proposes a novel approach to terrestrial IoT device positioning using LoRa or FSK modulation, where positioning is performed on the satellite, saving GPS bytes and reducing packet size. By utilizing Doppler effects and satellite motion, the method calculates device positions, forming a positioning equation set without interfering with communication tasks. Challenges in Doppler effect measurement and positioning are addressed with a fine-grained frequency shift measurement method using ZoomFFT and zero-padding FFT, and a non-linear equation set is solved to determine the positions of terrestrial nodes, with constraints added to ensure practical solutions. The framework is validated through simulations, demonstrating a significant reduction in payload length and positioning accuracy comparable to GPS. In ideal cases, the framework can locate nodes with errors less than 60 meters without any GPS information from data packets.

Low-bit quantization has become widespread for compressing image super-resolution (SR) models for edge deployment, which allows advanced SR models to enjoy compact low-bit parameters and efficient integer/bitwise constructions for storage compression and inference acceleration, respectively. However, it is notorious that low-bit quantization degrades the accuracy of SR models compared to their full-precision (FP) counterparts. Despite several efforts to alleviate the degradation, the transformer-based SR model still suffers severe degradation due to its distinctive activation distribution. In this work, we present a dual-stage low-bit post-training quantization (PTQ) method for image super-resolution, namely 2DQuant, which achieves efficient and accurate SR under low-bit quantization. The proposed method first investigates the weight and activation and finds that the distribution is characterized by coexisting symmetry and asymmetry, long tails. Specifically, we propose Distribution-Oriented Bound Initialization (DOBI), using different searching strategies to search a coarse bound for quantizers. To obtain refined quantizer parameters, we further propose Distillation Quantization Calibration (DQC), which employs a distillation approach to make the quantized model learn from its FP counterpart. Through extensive experiments on different bits and scaling factors, the performance of DOBI can reach the state-of-the-art (SOTA) while after stage two, our method surpasses existing PTQ in both metrics and visual effects. 2DQuant gains an increase in PSNR as high as 4.52dB on Set5 (x2) compared with SOTA when quantized to 2-bit and enjoys a 3.60x compression ratio and 5.08x speedup ratio. The code and models are available at https://github.com/Kai-Liu001/2DQuant.

This paper presents FLEX, a real-time scheduling framework that adaptively allocates limited machine attention (i.e., computing resources) among different spatial views (partitioned by camera facing directions) and sensory modalities (i.e., LiDAR and cameras) within multi-modal multi-view machine perception on resource-constrained embedded platforms. It is achieved through the effective wiring of two features: First, considering the heterogeneous and time-varying criticality among views and modalities within dynamic sensing contexts (i.e., object locations), we calibrate an “anytime” multi-modal perception pipeline that dynamically adjusts the modality fusion strategies of each view. Second, to optimize the GPU processing throughput with time guarantees, FLEX centers around an adaptive batch scheduling algorithm that intelligently groups consecutive asynchronous view inspection tasks based on the job sequence1generated from a non-preemptive EDF schedule to maximize a measure of system utility, with the runtime elastic fusion used as a subroutine. Temporal load balancing is maintained during the scheduling by always ensuring the sequential schedulability of future tasks in early batching decisions. We implement FLEX on NVIDIA Jetson Orin and conduct extensive experiments with a large-scale driving dataset. The results demonstrate the superiority of FLEX in improving perception quality and system throughput with time guarantees.1Following the real-time conventions, we use “job” to denote the instance of periodic tasks.

Considering limited on-device resources, current practices are attempting to deploy a system-level mixture-of-experts (MoE)-based foundation LLM shared by multiple mobile apps on a device to support mobile intelligence. However, mobile apps are hard to customize their services that require tuning adapters associated with the LLM using private in-app data. The difficulty arises due to both the limited on-device resources and the restricted control that apps have over the foundation LLM. To address this issue, in this work, we propose LiteMoE, a novel proxy submodel tuning framework that supports mobile apps to efficiently fine-tune customized adapters on devices using proxy submodels. The key technique behind LiteMoE is a post-training submodel extraction method, whereby without additional re-training, we can identify and reserve critical experts, match and merge moderate experts, to extract a lightweight and effective proxy submodel from the foundation LLM for a certain app. We implemented a prototype of LiteMoE and evaluated it over various MoE-based LLMs and mobile computing tasks. The results show that with LiteMoE, mobile apps are able to fine-tune customized adapters on resource-limited devices, achieving 12.7% accuracy improvement and 6.6× memory reduction compared with operating the original foundation LLM.

Network monitoring is essential for network management and optimization. In modern data centers, fluctuations in flow rates and network congestion events (e.g., microbursts) typically manifest on a microsecond timescale. However, the time granularity of network monitoring systems has not been refined correspondingly to efficiently capture these behaviors. Attaining the monitoring granularity at the microsecond scale can greatly facilitate network performance analysis and management, but poses considerable challenges regarding memory, bandwidth, and deployment costs. We propose μMon, a novel microsecond-level network monitoring system for data centers. The key of μMon is WaveSketch, an innovative algorithm that measures and compresses flow rate curves using in-dataplane wavelet transform. WaveSketch allows for more accurate characterization of application traffic patterns and aids in profiling transport algorithms. Furthermore, by combining the fine-grained flow rate measurements with network-collected congestion information, μMon can 'replay' congestion events to analyze their cause and impact. We evaluate μMon through testbed deployment and simulations at a granularity of 8.192 μs. The evaluation results demonstrate that μMon can achieve a 90% accuracy in microsecond-level rate measurements with an average of 5 Mbps bandwidth overhead per host. Additionally, it can capture 99% heavy congestion events with 31--82 Mbps bandwidth overhead per switch.

Spatial Crowdsourcing (SC) is a promising service that incentives workers to finish location-based tasks with high quality by providing rewards. Worker recruitment is a core issue in SC, for which most state-of-the-art algorithms focus on designing incentive mechanisms based on the existing SC worker pool. However, they may fail when the number of SC workers is not enough, especially for the new SC platforms. In recent years, social networks have been found to be helpful for worker recruitment by selecting seed workers to spread the task information so as to inspire more social users to participate, but how to select seed workers remains a challenge. Existing methods typically require numerous iterative searches leading to inefficiency in facing the big picture and failing to cope with dynamic environments.

Blockchain and smart contracts are one of the key technologies promoting Web 3.0. However, due to security considerations and consistency requirements, smart contracts currently only support simple and deterministic programs, which significantly hinders their deployment in intelligent Web 3.0 applications. To enhance smart contracts intelligence on the blockchain, we propose SMART, a plug-in smart contract framework that supports efficient AI model inference while being compatible with existing blockchains. To handle the high complexity of model inference, we propose an on-chain and off-chain joint execution model, which separates the SMART contract into two parts: the deterministic code still runs inside an on-chain virtual machine, while the complex model inference is offloaded to off-chain compute nodes. To solve the non-determinism brought by model inference, we leverage Trusted Execution Environments (TEEs) to endorse the integrity and correctness of the off-chain execution. We also design distributed attestation and secret key provisioning schemes to further enhance the system security and model privacy. We implement a SMART prototype and evaluate it on a popular Ethereum Virtual Machine (EVM)-based blockchain. Theoretical analysis and prototype evaluation show that SMART not only achieves the security goals of correctness, liveness, and model privacy, but also has approximately 5 orders of magnitude faster inference efficiency than existing on-chain solutions.

Federated learning is a new distributed machine learning framework, where numerous heterogeneous clients collaboratively train a model without sharing training data. In this work, we consider a practical and ubiquitous issue when deploying federated learning in mobile environments: intermittent client availability, where the set of eligible clients may change during the training process. Such intermittent client availability would seriously deteriorate the performance of the classical federated averaging algorithm (FedAvg). Thus, we propose a simple distributed nonconvex optimization algorithm, called federated latest averaging (FedLaAvg), which leverages the latest gradients of all clients, even when the clients are not available, to jointly update the global model in each iteration. Our theoretical analysis shows that FedLaAvg achieves guaranteed convergence and a sublinear speedup with respect to the total number of clients. We implement FedLaAvg along with several baselines and evaluate them over the benchmarking MNIST and Sentiment140 data sets. The evaluation results demonstrate that FedLaAvg achieves more stable training than FedAvg in both convex and nonconvex settings and reaches a sublinear speedup. Source code and online supplement are available at the IJOC GitHub site ( http://dx.doi.org/10.1287/ijoc.2022.0057.cd , https://github.com/INFORMSJoC/2022.0057 ). History: Accepted by Ram Ramesh, Area Editor for Data Science & Machine Leaning. Funding: This work was supported by the National Key R&D Program of China [Grant 2022ZD0119100], the National Natural Science Foundation of China (NSFC) [Grants 61972252, 61972254, 62072303, 62025204, 62132018, 62202296, and 62202297], the Alibaba Innovation Research (AIR) Program, and the Tencent Rhino Bird Key Research Project. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0057 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0057 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

At ByteDance, where we execute over a million Spark jobs and handle 500PB of shuffled data daily, ensuring resource efficiency is paramount for cost savings. However, achieving optimization of resource efficiency in large-scale production environments poses significant challenges. Drawing from our practical experiences, we have identified three key issues critical to addressing resource efficiency in real-world production settings: 1 slow I/Os leading to excessive CPU and memory idleness, 2 coarse-grained resource control causing wastage, and 3 sub-optimal job configurations resulting in low utilization. To tackle these issues, we propose a resource efficiency governance framework for Spark workloads. Specifically, 1 we devise the multi-mechanism shuffle services, including Enhanced External Shuffle Service (ESS) and Cloud Shuffle Service (CSS), where CSS employs a push-based approach to enhance I/O efficiency through sequential reading. 2 We modify the Spark configuration parameter protocol, allowing for fine-grained resource control by introducing several new parameters such as milliCores and memoryBurst, as well as supporting operators with additional spill modes. 3 We design a two-stage configuration autotuning method, comprising rule-based and algorithm-based tuning, providing more reliable Spark configuration optimizations. By deploying these techniques on millions of Spark jobs in production over the last two years, we have achieved over 22% CPU utilization increase, 5% memory utilization increase, and 10% shuffle block time ratio decrease, effectively saving millions of CPU cores and petabytes of memory daily.

Abstract In this article, we propose a mobile edge computing (MEC)‐related system named PD‐Gait, which can measure gait parameters of Parkinson's disease patients in a contactless and privacy‐preserving manner. We utilize inaudible acoustic signals and band‐pass filters to achieve privacy data protection in the physical layer. The proposed framework can be easily deployed in the mobile end of MEC, and hence release the edge server in cybersecurity attacks fighting. The gait parameters include stride cycle time length and moving speed, and hence providing an objective basis for the doctors' judgment. PD‐Gait utilizes acoustic signals in bands from 16 to 23 kHz to achieve device‐free sensing, which would release both doctors and patients from the tedious wearing process and psychological burden caused by traditional wearable devices. To achieve robust measurement, we propose a novel acoustic ranging method to avoid “broken tones” and “uneven peak distribution” in the received data. The corresponding ranging accuracy is 0.1 m. We also propose auto‐focus micro‐Doppler features to extract robust stride cycle time length, and can achieve an accuracy of 0.052 s. We deployed PD‐Gait in a brain hospital and collected data from 8 patients. The total walked distance is over 330 m. From the overall trend, our results are highly correlated with the doctor's judgment.

Airlines commonly need to take into consideration maximizing their profit while designing the hub-and-spoke network to obtain more market share and promote healthy development of aviation industry. Hence, in this article, we study the problem of multiple-allocation HUb and spoke network design for ROuting flight flows to maximize airline profit utility (HURO). That is, given a set of airport nodes, a set of flight flows with known origin positions and destination positions, finding a limited number of hub edges to transfer flows and determining routing allocation mode considering customer preference such that the overall transportation profit utility is maximized. To address HURO problem, we first consider a relaxed version of HURO (HURO-R for short). We prove that HURO-R falls into the realm of maximizing a submodular set function subject to a cardinality constraint, and propose an algorithm with a constant approximation ratio. Next, we design a two-level algorithm framework with a constant approximation ratio to address HURO. Besides, we consider variants of HURO, HURO-C and HURO-RU, and design approximation algorithms to address them. We conduct simulation experiments on standard dataset and field experiments to verify our theoretical findings. The results shows that our proposed algorithm can outperform other comparison algorithms by 75.28 percent.

Index structures are powerful tools for improving query performance and reducing disk access in database systems. Multi-dimensional indexes, in particular, are used to filter records effectively based on multiple attributes. Classical multi-dimensional index structures, such as KD-Tree, Quadtree, and R-Tree, have been widely used in modern databases. However, advancements in hardware and algorithms have led to the emergence of new types of multi-dimensional index structures. In this paper, we begin by reviewing classical multi-dimensional indexes. Next, we explore the approaches that leverage modern hardware features, such as Solid-State Drive, Non-Volatile Memory, Dynamic Random Access Memory, and Graphics Processing Unit, to improve the performance of multi-dimensional indexes in various aspects. Then, we investigate the novel work of multi-dimensional indexes that apply state-of-the-art machine learning techniques. Finally, we discuss the challenges and future research directions for multi-dimensional indexing methods.

Graph Contrastive Learning (GCL) with Graph Neural Networks (GNN) has emerged as a promising method for learning latent node representations in a self-supervised manner. Most of existing GCL methods employ random sampling for graph view augmentation and maximize the agreement of the node representations between the views. However, the random augmentation manner, which is likely to produce very similar graph view samplings, may easily result in incomplete nodal contextual information, thus weakening the discrimination of node representations. To this end, this paper proposes a novel trainable scheme from the perspective of node augmentation, which is theoretically proved to be injective and utilizes the subgraphs consisting of each node with its neighbors to enhance the distinguishability of nodal view. Notably, our proposed scheme tries to enrich node representations via a multi-scale contrastive training that integrates three different levels of training granularity, i.e., subgraph level, graph- and node-level contextual information. In particular, the subgraph-level objective between augmented and original node views is constructed to enhance the discrimination of node representations while graph- and node-level objectives with global and local information from the original graph are developed to improve the generalization ability of representations. Experiment results demonstrate that our framework outperforms existing state-of-the-art baselines and even surpasses several supervised counterparts on four real-world datasets for node classification.

Charger placement is an important problem in improving the quality of service in wireless rechargeable sensor networks. This paper studies the problem ofWireless ChArger PlacemeNt with Multiple (Directional) Antennas (WANDA). The problem is described as follows: given a set of wireless chargers equipped with multiple directional antennas and a set of wireless rechargeable sensors, determine the chargers' positions and orientations to maximize the overall charging utility. According to the relative positional relationship between the antennas, the problem is classified into Relative Orientation Fixed (WANDA-ROF) and Relative Orientation Unfixed (WANDA-ROU) situations. To address WANDA, we present a piecewise constant function to approximate the nonlinearity of charging power and propose an area discretization technique to reduce the infinite solution space to a limited one without performance loss. Then, we prove the monotonic submodularity of WANDA, and present a$\frac{1}{2}-\epsilon$approximation algorithm for the ROF situation and a$\frac{1}{2}-\epsilon$approximation algorithm for the ROU situation, all run in polynomial time. Finally, we conduct extensive simulation and experiments to show that our algorithms outperform comparison algorithms by at least 16% for ROF situation and 12% for ROU situation.

Wireless Power Transfer (WPT) has received more and more attention for its convenience and reliability. In this paper, we first propose the notion of omnidirectional charging. First, we consider the problem of detecting whether the target area achieves omnidirectional charging given a deterministic deployment of chargers. We use piecewise constant approximation and area discretization techniques to partition the target area and approximate charging power as constants. Next, we propose the Minimum Coverage Set extraction technique to design a fast detection algorithm. Second, we design a charger deployment scheme that satisfies omnidirectional charging. By placing the chargers at the triangle lattice points, we estimate the length of triangle lattice side length that satisfies omnidirectional charging, and derive the error bound with the optimal length. Third, we determine the probability that the target area achieves omnidirectional charging given a random deployment of chargers. We devise both analytical and numerical solutions for the problem with good accuracy. Finally, we conduct simulation and field experiments, and the results show that the running speed of our omnidirectional charging detection algorithm is at least$1\times$faster than comparison algorithms, and the consistency degree of our theoretical results and field experimental results is larger than$93.6 \%$.

Mobile crowdsensing (MCS) can promote data acquisition and sharing among mobile devices. Traditional MCS platforms are based on a triangular structure consisting of three roles: data requester, worker (i.e. , sensory data provider) and MCS platform. However, this centralized architecture suffers from poor reliability and difficulties in guaranteeing data quality and privacy, even provides unfair incentives for users. In this paper, we propose a blockchain-based MCS platform, namely BlockSense, to replace the traditional triangular architecture of MCS models by a decentralized paradigm. To achieve the goal of trustworthiness of BlockSense, we present a novel consensus protocol, namely Proof-of-Data (PoD), which leverages miners to conduct useful data quality validation work instead of “useless” hash calculation. Meanwhile, in order to preserve the privacy of the sensory data, we design a homomorphic data perturbation scheme, through which miners can verify data quality without knowing the contents of the data. We have implemented a prototype of BlockSense and conducted case studies on campus, collecting over 7,000 data from workers' mobile phones. Both simulations and real-world experiments show that BlockSense can not only improve system security, preserve data privacy and guarantee incentives fairness, but also achieve at least 5.6x faster than Ethereum smart contracts in verification efficiency.

This paper studies connectivity maintenance in uncertain networks under adversarial attack, where a defender conceals crucial links to prevent the largest connected component from being decomposed by an attacker. In contrast with its static counterpart, connectivity maintenance in uncertain networks involves additional probing on links to determine their existence. Therefore, by modeling an uncertain network as a random graph with each link associated with an existence probability and a probing cost, our goal is to design a defensive strategy for link selection that maximizes the expected size of the largest remaining connected component with the minimum expected probing cost, and moreover, the strategy should be independent of the attacking patterns. To this end, we first unravel the computational complexity of the problem by proving its NP-hardness, and then propose optimal defensive strategies based on dynamic programming and multi-objective optimization. Due to the prohibitive computational cost of optimality, two approximate defensive strategies are further designed to pursue decent performance with quasilinear complexity, in which the first one is a heuristic approach that quantifies the link vulnerability through an analogy from the degree centrality of a vertex in static networks to the connectivity weight of a link in uncertain networks, and the second one is an adaptive greedy policy incorporating the minimax rule from game theory, which minimizes the possible loss suffered by the defender in a worst-case scenario and has a constant approximation ratio. Extensive experiments on both synthetic and real-world network datasets under diverse attacking patterns demonstrate the superiority of the proposed strategies over baselines.

Truck-drone systems, wherein trucks carrying drones drive to pre-planned positions and then free drones equipped with cameras to monitor a known number of objects with reported positions, have been used for various scenarios. An object's quality of monitoring (QoM) by a camera is defined as a function of camera focal length and monitoring distance. Improving the QoM would help downstream tasks, including object detection and recognition. The monitoring utility is the fusion of all the QoMs of an object from multiple cameras. This paper optimizes theDeploymentOfTrucksAndDrones forObject monitoring (DOTADO) problem,i.e., deploying a truck-drone system, where each drone is equipped with a varifocal camera, to maximize the overall monitoring utility for all objects. Firstly, we model the hybrid system and define monitoring quality and utility. Then, we discretize the solution space into deployment strategies with performance bound. To select deployment strategies, we prove the submodularity of the problem and propose a two-level greedy algorithm with a bounded approximation ratio. Finally, we devise an optimal method to adjust the strategy for energy saving and communication improvement without losing monitoring utility. We perform both simulations and field experiments to verify the proposed framework.

Acoustic ranging has been viewed as a promising Human-Computer Interaction (HCI) technology in many scenarios, such as Augmented Reality (AR)/Virtual Reality (VR) and smart appliances. Most ranging systems with distributed devices undergo an extra calibration process to remove the timing errors. However, the calibration process needs user intervention. Furthermore, it should assume that the clock drifts are linear and stable, which is disabled within tens of minutes. In this paper, we introduce a self-calibrated acoustic ranging system that achieves sub-millimeter accuracy on distributed asynchronous devices. Based on our theoretical timing model, we precisely cancel both the system delay and the nonlinear clock drift with carefully designed Orthogonal Frequency-Division Multiplexing (OFDM) ranging signals. Our synchronization scheme achieves a timing accuracy of 1.9 microseconds, which allows us to build large-scale virtual acoustic arrays. Based on such a calibration scheme, our localization system achieves a ranging error of$\rm{0.39}~mm$within three meters in real-world experiments.

Mobile health (mHealth) applications, benefiting from mobile computing, have generated numerous mHealth data. However, they are dispersed across isolated devices, which hinders discovering insights underlying the aggregated data. Considering the online characteristics of mHealth, in this work, we present the first online dataVAluation andPricing mechanism, namely VAP, to incentive users to contribute mHealth data for machine learning (ML) tasks in mHealth systems. Under the Bayesian framework, we propose a new metric based on the concept of entropy to calculate data valuation during model training in an online manner. In proportion to the data valuation, we then determine payments as compensations for users to contribute their data. We formulate this pricing problem as a contextual multi-armed bandit with the goal of profit maximization and propose a new algorithm based on the characteristics of pricing. Furthermore, to tackle the budget constraint, we incorporate a two-stage multi-armed bandit with a knapsack method. We also extend VAP to advanced ML models by computing the entropy on the prediction space. Finally, we have evaluated VAP on two real-world mHealth data sets. Evaluation results show that VAP outperforms the state-of-the-art data valuation and pricing mechanisms in terms of computational complexity and extracted profit.

In the last decade, MIMO spatial multiplexing and distributed beamforming play a significant role in improving data throughput through cooperative transmission. It has been widely used in wireless communication, especially in 6G. However, the distributed uplink beamforming is still an open problem in highly dynamic environments. However, the proposed 6G technology represents the further integration of deep learning and wireless communication. In this article, we propose Argute Distributed Uplink Beamforming (ArguteDUB), which uses a feedback algorithm with an offline-trained deep learning model to implement highly dynamic distributed uplink beamforming for the Internet of Vehicles (IoV) in 6G. Specifically, each vehicle enables the base station (BS)/access point (AP) to separate different channel state information (CSI) by inserting orthogonal sequences into the sending data. The BS adopts deep learning to filter the noise and predict the beamforming weight to achieve phase synchronization. Unlike traditional distributed uplink beamforming, ArguteDUB can be adapted to the highly dynamic time-varying channels. The simple network structure ensures the fast response of ArguteDUB. In addition, we make ArguteDUB Orthogonal Frequency Division Multiplexing (OFDM) compatible so that it can be easily deployed in 6G networks. Our evaluation shows that ArguteDUB has an signal-to-noise ratio (SNR) gain of about 5 dB to 5.3 dB over the single vehicle transmission mode.

Spatial crowdsourcing, a human-centric paradigm for performing spatial tasks, has drawn rising attention. Task assignment and worker scheduling are basic issues in spatial crowdsourcing, which has a high demand on the timeliness of tasks and fairness among workers. In this article, we propose Utility-Fairness Index to evaluate the performance of the crowdsourcing platform and introduce the Fairness-Aware Task Planning problem that maximizes the utility while well-maintaining fairness during task assignment. We prove that its offline version is NP-hard, and clarify that there is no deterministic online algorithm with a constant competitive ratio for it. Two greedy-based online algorithms, Traversal Search and Advanced Nearest Neighbor are designed to solve the problem. We make optimization on running time with recurrence method to reach linear complexity, and use Gaussian Mixture Model to reveal the distribution regularity of tasks in these algorithms. Experiments on both synthetic and real datasets validate the efficiency and effectiveness of our algorithms.

Flow monitoring is widely applied in software-defined networks (SDNs) for monitoring network performance. Especially, detecting heavy hitters can prevent the Distributed Denial of Service (DDoS) attack. However, many existing approaches fall into one of two undesirable extremes: (i) inefficient collection where only accuracy is concerned in the method; (ii) sacrifice of accuracy due to fast detection. One practical problem with this is that it does not have the flexibility to adjust the monitoring strategy to the monitoring needs, making it difficult to meet different applications. To alleviate this problem, we propose our design of a novel flow monitoring framework that keeps the balance between accuracy and efficiency. It provides customized monitoring services for applications, where network resources can be saved, and the error rate can also be confined. In this paper, we present cReFeR, a three-step “compression Report-Feedback-Report” framework to monitor SDNs. The IP and the value compressor are specially designed to reduce the volume of flow statistics collection. This framework thus can achieve accuracy-ensured and resource-saving flow monitoring in SDNs. Theoretical analysis and simulated evaluation have proved the effectiveness of our solution. cReFeR keeps the error rate under 3% and reduces the amount of monitoring data more than 40%, which guarantees high efficiency compared with existing methods.

Consistent routing updates through Software-Defined Networking (SDN) can be difficult due to the asynchronous and distributed nature of the data plane. Recent studies have achieved consistent unicast routing updates. However, achieving consistent updates with drop-freeness and duplicate-freeness remains a challenge for multicast with fewer known results. This paper proposes a Novel Ordered Update Scheme called Nous, a novel approach that offers a comprehensive solution for consistently updating multicast routing based on SDN. To avoid duplicate entries, Nous configures the inport match field in the forwarding rules. Nous implements a dependency graph to schedule update operations dynamically. It also solves the Replace Operation Tree Migration Problem (ROTMP) using a greedy solution. To compare the greedy solution with the optimal solution, we employ the state-of-the-art mathematical programming solver Gurobi Optimizer 7.5 (for solving the optimization problem), Mininet 2.0, and Floodlight 1.2 (for simulation and comparison) to obtain a near-optimal solution. Simulation results show that using the greedy solution, Nous can usually achieve near-optimal solutions to the ROTMP with an average of fewer than 1.2 rounds and within 10 ms in different scenarios. This makes Nous the first ordered update scheme to guarantee two consistent states simultaneously.

Machine learning (ML) models have been deployed in mobile networks to deal with massive data from different layers to enable automated network management. To overcome high communication cost and severe privacy concerns of centralized ML, federated learning (FL) has been proposed to achieve distributed ML among numerous networked devices. While the computation and communication limitation has been widely studied, the impact of limited storage of mobile devices on the performance of FL is still not explored. Without an effective data selection policy to filter the massive streaming networked data on devices, classical FL can suffer from much longer model training time ($4\times$) and dramatic inference accuracy reduction ($7\%$), observed in our experiments. In this work, we take the first step to consider the online data selection for FL with limited on-device storage. We first define a new data valuation metric for data selection in FL with theoretical guarantee for simultaneously accelerating model convergence and enhancing final accuracy. We further design ODE, an Online Data sElection framework for FL, to coordinate networked devices to store valuable data samples collaboratively. Experimental results on one industrial dataset and three public datasets show the remarkable advantages of ODE over the state-of-the-art approaches. Particularly, on the industrial dataset, ODE achieves as high as$2.5\times$speedup of training time and$6\%$increase in final accuracy, and is robust to various factors in practical environments.

Nowadays, there exists a lot of cross-region data transmission demand on the cloud. It is promising to use serverless computing for data compressing to save the total data size. However, it is challenging to estimate the data transmission time and monetary cost with serverless compression. In addition, minimizing the data transmission cost is non-trivial due to the enormous parameter space. This paper focuses on this problem and makes the following contributions: 1) We propose empirical data transmission time and monetary cost models based on serverless compression. It can also predict compression information, e.g., ratio and speed using chunk sampling and machine learning techniques. 2) For single-task cloud data transmission, we propose two efficient parameter search methods based on Sequential Quadratic Programming (SQP) and Eliminate then Divide and Conquer (EDC) with proven error upper bounds. Besides, we propose a parameter fine-tuning strategy to deal with transmission bandwidth variance. 3) Furthermore, for multi-task scenarios, a parameter search method based on dynamic programming and numerical computation is proposed. We have implemented the system called Fluid-Shuttle, which includes straggler optimization, cache optimization, and the autoscaling decompression mechanism. Finally, we evaluate the performance of Fluid-Shuttle with various workloads and applications on the real-world AWS serverless computing platform. Experimental results show that the proposed approach can improve the parameter search efficiency by over$3\times $compared with the state-of-art methods and achieves better parameter quality. In addition, our approach achieves higher time efficiency and lower monetary cost compared with competing cloud data transmission approaches.

RDMA over Converged Ethernet (RoCEv2) has been widely deployed to data centers (DCs) for its better compatibility with Ethernet/IP than Infiniband (IB). As cross-DC applications emerge, they also demand high throughput, low latency, and lossless network for cross-DC data transmission. However, RoCEv2’s underlying lossless mechanism Priority-based Flow Control (PFC) cannot fit into the long-haul transmission scenario and degrades the performance of RoCEv2. PFC is myopic and only considers queue length to pause upstream senders, which leads to large queueing delay. This paper proposes Bifrost, a downstream-driven lossless flow control that supports long distance cross-DC data transmission. Bifrost uses virtual incoming packets, which indicates the upper bound of in-flight packets, together with buffered packets to control the flow rate. It minimizes the buffer space requirement to one-hop bandwidth delay product (BDP) and achieves low one-way latency. Moreover, we extend Bifrost and propose BifrostX, to accommodate the multi-priority queue of the current switch implementation. BifrostX enables flow control for each queue separately while maintaining low buffer reservation, no throughput loss, and no packet loss. Real-world experiments are conducted with prototype switches and 80 kilometers cables. Evaluations demonstrate that compared to PFC, Bifrost reduces average/tail flow completion time (FCT) of inter-DC flows by up to 22.5%/42.0%, respectively. Bifrost is compatible with existing infrastructure and can support distance of thousands of kilometers.

In-air gesture control extends a touch screen and enables contactless interaction, thus has become a popular research direction in the past few years. Prior work has implemented this functionality based on cameras, acoustic signals, and Wi-Fi via existing hardware on commercial devices. However, these methods have low user acceptance. Solutions based on cameras and acoustic signals raise privacy concerns, while WiFi-based solutions are vulnerable to background noise. As a result, these methods are not commercialized and recent flagship smartphones have implemented in-air gesture recognition by adding extra hardware on-board, such as mmWave radar and depth camera. The question is, can we support in-air gesture control on legacy devices without any hardware modifications? To answer this question, in this work, we propose, an in-air gesture recognition system leveraging the screen and ambient light sensor (ALS), which are ordinary modalities on mobile devices. For the transmitter side, we design a screen display mechanism to embed spatial information and preserve the viewing experience; for the receiver side, we develop a framework to recognize gestures from low-quality ALS readings. We implement and evaluate on both a tablet and several smartphones. Results show that can recognize$9$types of frequently used in-air gestures with an average accuracy of$96.1\%$.

Finding special items in data streams, like heavy hitters, top-$k$items, and persistent items, has always been a hot topic in the field of network measurement. While data streams nowadays are usually high-dimensional, most prior works optimize data structures to accurately find special items according to a certain primary dimension and yield little insight into the correlations between dimensions, where the dimension can be a single data dimension or a combination of multiple data dimensions. Therefore, we propose to find special quadratic elements in data streams to reveal the close correlations between the primary and secondary dimensions. Here, both the primary and secondary dimensions are selected according to specific application purposes. Based on the special items mentioned above, we extend our problem to three applications related to heavy hitters, top-$k$, and persistent items, and design a generic framework DUET to process them. We analyze the error bound of our algorithm theoretically and conduct extensive experiments on four publicly available data sets. Our experimental results show that DUET can achieve 3.5 times higher throughput and three orders of magnitude lower average relative error than cutting-edge algorithms. Moreover, we propose an optimized framework based on DUET, namely O-DUET, to further improve the estimation accuracy. We also discuss a hardware-version DUET and deploy it on Tofino.

Spectrum sensing plays a crucial role in spectrum monitoring and management. However, due to the expensive cost of high-speed ADCs, wideband spectrum sensing is a long-standing challenge. In this paper, we present how to transform Ultra-wideband (UWB) devices into a spectrum sensor which can provide wideband spectrum monitoring at a low cost. Compared with the expensive high-speed ADCs which cost at least hundreds of dollars, a UWB device is only several dollars. As the low-cost UWB technology is not originally designed for spectrum sensing, we address the inherent limitations of low-cost devices such as limited memory, low SPI speed and low accuracy, and show how to obtain spectrum occupancy information from the noisy and spurious UWB channel impulse response. In this paper, we present, which not only can give accurate channel occupancy information, but also can precisely estimate the signal power and bandwidth. can also detect fleeting radar signals. We implement and perform extensive evaluations with both controlled experiments and field tests. Results show that can sense up to$900$MHz bandwidth with frequency range from 0.5GHz to 7GHz, and the power estimation error is less than$3$dB. can also accurately detect busy 5G channels and can classify the encrypted traffic from the channel usage patterns. We believe that provides a new paradigm for low-cost wideband spectrum sensing, which is critical for large-scale fine-grained spectrum monitoring.

The quality of the video stream is the key to neural network-based video analytics. However, low-quality video is inevitably collected by existing surveillance systems because of poor-quality cameras or over-compressed/pruned video streaming protocols, e.g., as a result of upstream bandwidth limit. To address this issue, existing studies use quality enhancers (e.g., neural super-resolution) to improve the quality of videos (e.g., resolution) and eventually ensure inference accuracy. Nevertheless, directly applying quality enhancers does not work in practice because it will introduce unacceptable latency. In this paper, we present AccDecoder, a novel accelerated decoder for real-time and neural-enhanced video analytics, selects a few frames adaptively via Deep Reinforcement Learning (DRL) to enhance the quality and inference then reuse on the unselected ones. Next, we extend AccDecoder to AccDecoder$+$by formulating the resolution-involved Markov decision process (MDP) to achieve resolution adaptation; it aims to trade accuracy and latency corresponding under various video resolutions. Proved by experiments, AccDecoder provides efficient inference capability via filtering important frames using DRL for DNN-based inference and reusing the results for the other frames via extracting the reference relationship among frames and blocks, which contributes 6-21% accuracy improvement and a latency reduction of 20-80% than baselines. Compared with AccDecoder, AccDecoder$+$achieves an additional 2-7% accuracy improvement.

Cloud services have shifted from monolithic designs to microservices running on cloud-native infrastructure with monitoring systems to ensure service level agreements (SLAs). However, traditional monitoring systems no longer meet the demands of cloud-native monitoring. In Alibaba’s “double eleven” shopping festival, it is observed that the monitor occupies resources of the monitored infrastructure and even disrupts services. In this paper, we propose a novel monitoring system named for cloud-native monitoring. achieves zero overhead in collecting raw metrics using one-sided remote direct memory access (RDMA) and remedies network congestion by adopting a receiver-driven flow control scheme. also features a priority queue mechanism to meet different quality of service requirements and an efficient batch processing design to relieve CPU occupation. has been deployed and evaluated in four different clusters with heterogeneous RDMA NIC devices and architectures in Alibaba Cloud. Results show that achieves no CPU occupation at the monitored host and supports$1\sim10k$hosts with$0.1\sim1s$sampling interval using a single thread for network I/O. significantly relieves the incast issue and maintains$80\sim95\%$of bandwidth utilization in several clusters when monitoring$1k$hosts. also ensures services with high priority accomplish collecting metrics earlier than low priority ones by at least$400 \mu s$when monitoring$1k$hosts.

The approximate membership query (AMQ) data structure is a kind of space-efficient probabilistic data structure. It can approximately indicate whether an element exists in a set. The AMQ data structure has been widely used in network measurements, network security, network caching,etc. Resizing is an extensively utilized operation of the AMQ data structure, but it can lead to system performance degradation. We summarize two main problems that lead to such degradation. Specifically, one of them is that the resizing operation can block other operations, while the other one is that the throughput of AMQ structures will deteriorate after multiple resizing operations due to more computation cost. However, existing related work cannot alleviate both of them. Therefore, we propose a novel AMQ data structure called bamboo filters, which can alleviate the two problems simultaneously. Bamboo filters can insert, look up, and delete an element in constant time. They can also dynamically resize in a fine-grained way. Furthermore, we propose space utilization adaptive bamboo filters that adaptively trigger resizing operations according to the space utilization, thereby achieving lower average memory consumption. Experimental results show that our scheme significantly outperforms state-of-the-art work. Especially, bamboo filters achieve 2.12$\times$lookup throughput of the logarithmic dynamic cuckoo filter.

Multipath TCP (MPTCP) is a burgeoning transport protocol which enables the server to transmit the traffic across multiple network interfaces in parallel. Classic MPTCPs have good friendliness and practicality such as relatively low overhead, but are hard to achieve consistent high-throughput and adaptability, especially for the ability to flexibly balance the congestion among different subpaths. In contrast, learning-based MPTCPs can essentially achieve consistent high-throughput and adaptability, but have poor friendliness and practicality. In this paper, we proposed MPLibra, a combined multipath congestion control framework that can complement the advantages of classic MPTCPs and learning-based MPTCPs together. MPLibra periodically leverages both classic MPTCPs and learning-based MPTCPs to make decisions and select the better one based on real-time network feedbacks. Extensive simulations on NS3 show that MPLibra can achieve good performance and outperform state-of-the-art MPTCPs under different network conditions. MPLibra improves the throughput by 40.5% and reduces the file download time by 29.94% compared with LIA, achieves good friendliness and balances congestion timely. What’s more, on the basis of MPLibra, we propose MPLibra+ which adds a safety module and an optimized packet scheduler and is the upgrade version of MPLibra. MPLibra+ has better ability to cope with untrained network environment and achieve better performance on heterogeneous scenarios compared with MPLibra.

Data centers heavily rely on scale-out architectures like fat-tree, BCube and VL2 to accommodate a large number of commodity servers. Since the traditional electrical network is demand-oblivious and cannot perfectly respond to the bursty traffic generated by big data applications, a growing trend is to design demand-aware topologies via introducing the converters with adjustable optical links, instead of adding more wiring links. However, little is known today about how to fully exploit the potential of the flexibility from the converters: the joint optimization on adjusting the optical links inside the converters and the routing in the whole network remains algorithmically challenging. In this paper, we design a set of customized converters for Diamond, VL2 and BCube topologies and initiate the optimization study in hybrid switching data center networks. As a case study, we introduce demand-aware load balancing problem (DLBP), i.e., a joint optimization on the physical layer (how the optical links interconnect inside the converter) and the network layer (how to determine the route especially for elephant flows in the whole network). We prove that DLBP is not only NP-hard, but also$\rho $-inapproximation. Accordingly, we design two algorithms: the first one is an intuitive greedy algorithm and the second one uses reinforcement learning to improve upon the solution of the first one. Trace-driven evaluations show that our algorithms can reduce the traffic congestion by 12% on average.

Internet of Vehicles (IoV) has become an indispensable technology to bridge vehicles, persons, and infrastructures and is promising to make our cities smarter and more connected. It enables vehicles to exchange vehicular data (e.g., GPS, sensors, and brakes) with different entities nearby. However, sharing these vehicular data over the air raises concerns about identity privacy leakage. Besides, the centralized architecture adopted in existing IoV systems is fragile to single point-of-failure and malicious attacks. With the emergence of blockchain technology, there is the chance to solve these problems due to its features of being tamper-proof, traceability, and decentralization. In this article, we propose a privacy-preserving vehicular data sharing framework based on blockchain. In particular, we design an anonymous and auditable data sharing scheme using Zero-Knowledge Proof (ZKP) technology so as to protect the identity privacy of vehicles while preserving the vehicular data auditability for Trusted Authorities (TAs). In response to high mobility of vehicles, we design an efficient multi-sharding protocol to decrease blockchain communication costs without compromising the blockchain security. We implement a prototype of our framework and conduct extensive experiments and simulations on it. Evaluation and analysis results indicate that our framework can not only strengthen system security and data privacy but also reduce communication complexity by \(O(\frac{n\sqrt {m}}{m^2})\) times compared to existing sharding protocols.

Spatial crowdsourcing is an increasingly popular category in the era of mobile Internet and sharing economy, where tasks have spatio-temporal constraints and must be completed at specific locations. In this paper, we focus ontheMulti-ObjectiveSpatio-Temporal task assignment (MOST) problemconsidering the worker heterogeneity in spatial crowdsourcing and model it as a combinatorial multi-objective optimization (MOO) problem with the goals of maximizing the overall task completion rate and minimizing the average task time cost. Finding the optimal global assignment turns out to be intractable since it does not simply imply optimality for an individual worker, as a typical nearest-neighbor heuristic generally does not render a satisfactory result. We prove that the problem is NP-hard. Subsequently, we formulate an efficient algorithm for the MOST problem —Task Clustering basedMixedPriority Queue Scheduling (TAMP). First, we improve the spectral clustering algorithm to evenly divide the task network into different subdomains according to tasks' geographical locations, considering the task clustering phenomena in real scenarios. We then design a mixed priority queue strategy considering the geographical influence and temporal urgency, to schedule workers finishing tasks in sequence. Experiments on synthetic and real datasets demonstrate the efficiency of our solution over other methods.

Travel route recommendation is an important part of electronic tour guides and map applications. It aims to recommend a sequence of points of interest (POIs) to users based on their interests. The variety of users' historical records and their requirements makes the problem challenging and most existing works fail to satisfy these two aspects at the same time. In this paper, we propose an encoder-decoder-based travel route recommendation framework, to help electronic tourist guide applications better recommend routes for their users. The framework makes accurate and flexible route recommendations by combining encoder-decoder structure with grid beam search. We make feature extraction and feature completion with domain knowledge and matrix factorization methods. Then, an encoder-decoder structure with a dual bi-directional LSTM encoder is proposed as a basis for route generation. Finally, we select the routes by grid beam search algorithm to give efficient recommendations. Multiple explicit requirements can be supported in our model, including unavailable POIs, mandatory POIs, restricted sequence length, and dynamic route revision. Experiments on eight real-world datasets show that our model achieves a 9.8% improvement in performance, compared with the state-of-the-art and supports more explicit requirements.

Reactive congestion control (RCC) protocols have undergone decades of evolution, where senders first send data packets and then back off when congestion occurs. Recently, there has been a surge of interest in proactive congestion control (PCC) that allocates bandwidth before transmission. Despite its potential, we found that there are certain scenarios where PCC may fall short. In this paper, we aim to provide a comprehensive understanding of PCC and motivate further exploration of this area. We conduct case studies and leverage NS3 simulations to compare state-of-the-art PCC with RCCs, delving into the real dilemma of PCC.

With increasing concern over data privacy, split learning has become a widely used distributed machine learning paradigm in practice, where two participants (namely the non-label party and the label party) own raw features and raw labels respectively, and jointly train a model. Although no raw data is communicated between the two parties during model training, several works have demonstrated that data privacy, especially label privacy, is still vulnerable in split learning, and have proposed several defense algorithms against label attacks. However, the theoretical guarantee on the privacy preservation of these algorithms is limited. In this work, we propose a novel Private Split Learning Framework (PSLF). In PSLF, the label party shares only the gradients computed by flipped labels with the non-label party, which improves privacy preservation on raw labels, and meanwhile, we further design an extra sub-model from true labels to improve prediction accuracy. We also design a Flipped Multi-Label Generation mechanism (FMLG) based on randomized response for the label party to generate flipped labels. FMLG is proven differentially private and the label party could make a trade-off between privacy and utility by setting the DP budget. In addition, we design an upsampling method to further protect the labels against some existing attacks. We have evaluated PSLF over real-world datasets to demonstrate its effectiveness in protecting label privacy and achieving promising prediction accuracy.

While learning to rank (LTR) is widely employed in web searches to prioritize pertinent webpages from the retrieved contents based on input queries, traditional LTR models stumble over two principal stumbling blocks leading to subpar performance: 1) the lack of well-annotated query-webpage pairs with ranking scores to cover search queries of various popularity, debilitating their coverage of search queries across the popularity spectrum, and 2) ill-trained models that are incapable of inducing generalized representations for LTR, culminating in overfitting. To tackle the above challenges, we proposed a Generative Semi-Supervised Pre-trained (GS2P) Learning to Rank model. Specifically, GS2P first generates pseudo-labels for the unlabeled samples using tree-based LTR models after a series of co-training procedures, then learns the representations of query-webpage pairs with self-attentive transformers via both discriminative (LTR) and generative (denoising autoencoding for reconstruction) losses. Finally, GS2P boosts the performance of LTR through incorporating Random Fourier Features to over-parameterize the models into “interpolating regime”, so as to enjoy the further descent of generalization errors with learned representations. We conduct extensive offline experiments on a publicly available dataset and a real-world dataset collected from a large-scale search engine. The results show that GS2P can achieve the best performance on both datasets, compared to baselines. We also deploy GS2P at a large-scale web search engine with realistic traffic, where we can still observe significant improvement in real-world applications. GS2P performs consistently in both online and offline experiments.

Deep learning models rely on highly optimized tensor libraries for efficient inference on heterogeneous hardware. Current deep compilers typically predetermine layouts of tensors and then optimize loops of operators. However, such unidirectional and one-off workflow strictly separates graph-level optimization and operator-level optimization into different system layers, missing opportunities for unified tuning.

Satellite internet (Sat-Net) enables high-speed con-nectivity with low latency and extensive coverage. However, it faces challenges related to data security and reliable networking. Blockchain technology, with its decentralized and tamper-proof nature, offers a promising solution to these challenges. However, applying blockchain to the Sat-Net topology is difficult due to its highly dynamic structure, which results in restricted communication periods between satellites and ground stations. To address these issues and adapt to the Sat-Net topology, we propose a solution called RelSharding, which utilizes relayer satellites to relieve concurrent transmission pressure of ground stations by collecting and transmitting data transactions from client satellites. Through caching block headers and Merkle trees in relayer satellites, RelSharding can also reduce authentication latency between satellites utilizing simplified payment verification (SPV). Conducted on the modified SimBlock framework, our experiments indicate that RelSharding can enhance throughput up to 45x and decrease latency by 292x compared to existing blockchain solutions.

Trend forecasting is vital to optimize the workload performance. It becomes even more urgent with an increasing number of applications and database configurations. However, DBAs mainly target at historical workloads and may give suboptimal configuration advice when the workload trends have changed. Although there are some studies on trend forecasting, they have several limitations. First, they mainly predict the changes of query numbers, which do not combine other critical factors (e.g., disk utilization) and cannot fully reflect the future workload trends. Besides, there are numerous queries in the workloads and exact clustering algorithms like K-means cannot effectively merge similar queries which contain noises like time shifts. Second, basic machine learning models like RNN may have relatively low prediction accuracy on complex workloads (e.g., no cycles but random bursts). Third, real-world workloads may have diverse patterns, while previous models cannot efficiently and reliably predict for all the different workload patterns.To address these challenges, we propose a trend forecasting system (DBAugur) that utilizes adversarial neural networks to predict the trends of different workloads. First, DBAugur collects the important features (e.g., queries, resource metrics) to characterize workloads, and reduces the number of involved queries by separately merging similar queries based on the SQL semantics and trend patterns. Second, DBAugur utilizes Generative Adversarial Networks (GANs) to capture the latent patterns, correlations between different metrics, and occasional bursts within the complicated and time-varying workloads. Moreover, we further propose a time-sensitive ensemble algorithm that takes advantage of various machine learning models (e.g., generative models, convolutional models, feed-forward models) to accommodate the various workload patterns. The experimental results show that DBAugur outperformed state-of-the-art methods on various real-world workloads.

With the growing demand for global trade transportation, the shipping container market has gained an increasingly important position. As a key issue of the market, container pricing is regarded as an important indicator to adjust the market supply and demand as well as the revenue of liner enterprises. Although various methods aimed at increasing enterprise revenue, such as expert pricing and dynamic pricing, have been proposed by industry and academia in recent years, these approaches rarely yield worst-case performance guarantee for the double-sided online scenarios of commodities and buyers.To cater to the double-sided online scenario and provide theoretical performance guarantee, we propose an online learning-based pricing framework named Balancing Inventory and Revenue with -chasing Decider (BIRD). BIRD determines container price by combining advantages of given multiple online pricing strategies. We utilize a strategy selector A to select a proper target strategy and use an ϵ-chasing decider ${{{\mathfrak{D}}}^{Cha\operatorname{s} ing}}$ to determine the price. BIRD is proven to combine the advantages of multiple online pricing strategies to achieve the performance close to the posterior optimal strategy for any sequence of online buyers on realistic sales platforms with inventory limitation. BIRD is proved to yield a vanishing regret for the online posted pricing problem with the features of limited inventory and multi-unit demand. Based on the historical data provided by COSCO, one of the largest liner enterprises in the world, we experimentally demonstrate the effectiveness of the proposed algorithm.

Estimating per-flow cardinality from high-speed data streams has many applications such as anomaly detection and resource allocation. Yet despite tracking single flow cardinality with approximation algorithms offered, there remain algorithmical challenges for monitoring multi-flows especially under unbalanced cardinality distribution: existing methods adopt a uniform sketch layout and incur a large memory footprint to achieve high accuracy. Furthermore, they are hard to implement in the compact hardware used for line-rate processing.In this paper, we propose Couper, a memory-efficient measurement framework that can estimate cardinality for multi-flows under unbalanced cardinality distribution. We propose a two-layer structure based on a classic coupon collector’s principle, where numerous mice flows are confined to the first layer and only the potential elephant flows are allowed to enter the second layer. Our two-layer structure can better fit the unbalanced cardinality distribution in practice and achieve much higher memory efficiency. We implement Couper in both software and hardware. Extensive evaluation under real-world and synthetic data traces show more than 20× improvements in terms of memory-efficiency compared to state-of-the-art.

Remote Direct Memory Access has been widely adopted in distributed storage systems. However, it only supports unicast operations, which degrades the performance significantly for data replication because of bandwidth waste and CPU overhead. To address the problem, we propose MC-RDMA, a distributed and reliable multicast RDMA. It is compatible with existing unicast RDMA but supports lazy packet replication with reliable RDMA multicasting. The key idea of MC-RDMA is utilizing in-network programmable switches to build a NIC-transparent reliable multicast protocol for RDMA. MC-RDMA combines the address information of the IP and RoCEv2 into a sender-initialized multicast routing protocol. Besides, it synchronizes the hardware transmission states of multiple receivers by merging ACKs and NAKs. To verify the effectiveness of MC-RDMA, we implement it with Mellanox ConnectX-6 commodity RNICs and Intel Tofino P4 programmable switches. Experimental results show that MC-RDMA can double the sender bandwidth utilization and reduce the CPU overhead significantly compared to unicast-based RDMA replications. Moreover, it reduces the storage request latency by -30% with realistic workloads and decreases the training time by -50% in the distributed training system.

RDMA over Converged Ethernet (RoCEv2) has been widely deployed to data centers (DCs) for its better compatibility with Ethernet/IP than Infiniband (IB). As cross-DC applications emerge, they also demand high throughput, low latency, and lossless network for cross-DC data transmission. However, RoCEv2's underlying lossless mechanism Priority-based Flow Control (PFC) cannot fit into the long-haul transmission scenario and degrades the performance of RoCEv2. PFC is myopic and only considers queue length to pause upstream senders, which leads to large queueing delay. This paper proposes Bifrost, a downstream-driven lossless flow control that supports long distance cross-DC data transmission. Bifrost uses virtual incoming packets, which indicates the upper bound of in-flight packets, together with buffered packets to control the flow rate. It minimizes the buffer space requirement to one-hop bandwidth delay product (BDP) and achieves low one-way latency. Real-world experiments are conducted with prototype switches and 80 kilometers cables. Evaluations demonstrate that compared to PFC, Bifrost reduces average/tail flow completion time (FCT) of inter-DC flows by up to 22.5%/42.0%, respectively. Bifrost is compatible with existing infrastructure and can support distance of thousands of kilometers.

As the emergence of recently popular large language model, distributed training (DT) optimizes the performance via using different parallelization strategies, resource schedulers and advanced compression techniques. Meanwhile, a promising acceleration primitive, In-Network Aggregation (INA), offloads the gradient aggregation to programmable switches to further reduce the communication overhead using the switch memory. However, to the best of our knowledge, how to identify the performance bottleneck in real time remains challenging. In this paper, we build Argus, a performance bottleneck monitoring framework for INA. Argus implements an aggregation digest extracting mechanism for real-time monitoring of DT jobs at the multi-tenant, multi-rack clusters. Argus models aggregation to identify performance bottlenecks in aggregation, which assists the scheduler in deciding the resource allocation. Extensive evaluation and prototype implementation show that Argus provides real-time and packet-level aggregation monitoring for identifying bottlenecks in INA with minimal performance overhead.

Federated learning (FL) in large-scale wireless networks is implemented in a hierarchical way by introducing edge servers as relays between the cloud server and devices, where devices are dispersed within multiple clusters coordinated by edges. However, the devices are usually mobile users with unpredictable mobile trajectories, whose effects on the model training process are still less studied. In this work, we propose a new MobIlity-Driven feDerated LEarning framework, namely MIDDLE in wireless networks, which can relieve unbalanced and biased model updates by leveraging the new model aggregation opportunities on mobile devices due to their mobility across edges. Specifically, mobile devices can have different models while traversing across edges, and adequately aggregate these models on the device. By theoretical analysis, we can show that this on-device model aggregation can reduce the bias of model updating on edges and cloud, and then accelerate the convergence of model training in FL. Then, we define a model similarity utility to measure the difference in gradient updates among various models, which guides the adaptive on-device model aggregation and in-edge device selection to facilitate the comprehensive information sharing between edges. Extensive experiment results validate that MIDDLE can achieve 1.51 × −6.85 × speedup on the model training, compared with the state-of-the-art model training approaches in hierarchical FL.

Social networks are widely recognized as highly effective information sources for social recommendation services. However, previous social recommendation methods assumed that a user’s preference factor and social trust factor shared a common latent feature space. Additionally, few studies have explored the incorporation of social information into the item domain for recommendations. To address these gaps, we propose ExpoEv, a deep collaborative filtering recommendation model that integrates social exposure based on feature evolution for social recommendation services. Specifically, we propose a social exposure module for both user and item domains that considers the number of items that a user’s social friends interact with. Furthermore, we introduce a feature evolution component that enables the incorporation of social exposure information with social trust and attribute factors in the context of social recommendation services. Experiments demonstrate the effectiveness of our model in the quality of recommendation service.

Automated bidding, an emerging intelligent decision-making paradigm powered by machine learning, has become popular in online advertising. Advertisers in automated bidding evaluate the cumulative utilities and have private financial constraints over multiple ad auctions in a long-term period. Based on these distinct features, we consider a new ad auction model for automated bidding: the values of advertisers are public while the financial constraints, such as budget and return on investment (ROI) rate, are private types. We derive the truthfulness conditions with respect to private constraints for this multi-dimensional setting, and demonstrate any feasible allocation rule could be equivalently reduced to a series of non-decreasing functions on budget. However, the resulted allocation mapped from these non-decreasing functions generally follows an irregular shape, making it difficult to obtain a closed-form expression for the auction objective. To overcome this design difficulty, we propose a family of truthful automated bidding auction with personalized rank scores, similar to the Generalized Second-Price (GSP) auction. The intuition behind our design is to leverage personalized rank scores as the criteria to allocate items, and compute a critical ROI to transforms the constraints on budget to the same dimension as ROI. The experimental results demonstrate that the proposed auction mechanism outperforms the widely used ad auctions, such as first-price auction and second-price auction, in various automated bidding environments.

Low-power wide-area networks (LPWAN) evolve rapidly with advanced communication primitives (e.g., coding, modulation) being continuously invented. This rapid iteration on LPWAN, however, forms a communication barrier between legacy wireless sensor nodes deployed years ago (e.g., ZigBee-based sensor node) with their latest competitor running a different communication protocol (e.g., LoRa-based IoT node): they work on the same frequency band but share different MAC- and PHY-layer regulations and thus cannot talk to each other directly. To break this barrier, we propose LigBee, a cross-technology communication (CTC) solution that enables symbol-level communication from the latest LPWAN LoRa node to legacy ZIGBEE node. We have implemented LigBee on both software-defined radios and commercial-off-the-shelf (COTS) LoRa and ZigBee nodes, and demonstrated that LigBee builds a reliable CTC link from LoRa node to ZigBee node on both platforms. Our experimental results show that i) LigBee achieves a bit error rate (BER) in the order of 10−3with 70 ∼ 80% frame reception ratio (FRR), ii) the range of LigBee link is over 300m, which is 6 ∼ 7.5× the typical range of legacy ZigBee and state-of-the-art solution, and iii) the throughput of LigBee link is maintained on the order of kbps, which is close to the LoRa’s throughput.

Remote Procedure Call (RPC) is widely used in distributed systems and it usually needs to serialize data before transmission. Serialization accounts for a large proportion of the overhead in RPC and becomes a bottleneck for RPC communications. Because the size of the output serialized message cannot be predicted in advance, there could be multiple memory reallocations and copies in typical serialization libraries (e.g., FlatBuffers), which dominates the overhead. We propose the novel serialization library, zFlatBuffers, to eliminate these avoidable copies during the serialization process and realize zero copy during communication. Unlike the typical serialization library, FlatBuffers, the message generated by zFlatBuffers consists of multiple non-contiguous buffers due to its zero-copy nature. Moreover, we integrate zFlatBuffers with RDMA-based RPC systems. For RDMA Unreliable Datagram, we modify the message buffer of eRPC to enable it to transmit messages composed of multiple buffers. We also build the zRPC system based on RDMA Reliable Connection, which transmits the zFlatBuffers message by the scatter/gather function. Compared to the original FlatBuffers, zFlatBuffers improves the throughput of eRPC and zRPC by 11.2%-33.7% and 5.8%-53.6%, respectively.

Vertical federated learning (VFL) allows multiple clients with misaligned feature spaces to collaboratively accomplish the global model training. Applying VFL to high stakes decision scenarios greatly requires model interpretation for decision reliability and diagnosis. However, the feature discrepancy in VFL raises new issues for model interpretation in distributed setting: one is from the local-global perspective, where the local importance of features is not equal to the global importance; and the other is from the local-local perspective, where information asymmetry among clients causes difficulty in identifying overlapped features. In this work, we propose a new distributed Model Interpretation method for Vertical Federated Learning with feature discrepancy, namely MI-VFL. In particular, to deal with the local-global discrepancy, MI-VFL leverages the law of total probability to adjust the local importance of features and ensures the completeness of the selected features using adversarial game. To handle the local-local discrepancy, MI-VFL builds a federated adversarial learning model to efficiently identify the overlapped features once, rather than performing client-to-client intersections multiple times. We extensively evaluate MI-VFL on six synthetic datasets and five real-world datasets. The evaluation results reveal that MI-VFL can accurately identify the important features, suppress the overlapped features, and thus improve the model performance.

Weighted congestion control aims to provide end-to-end differentiated bandwidth allocation. MulTCP and EWTCP are two closely related schemes for this purpose and they both want to achieve weighted proportionality through modifying AIMD behaviors. In this paper, we revisit the performance of MulTCP and EWTCP in terms of weighted proportionality. Through testbed experiments, we reveal a lot of counter-intuitive phenomena for achieving weighted proportionality — the switch buffer size, propagation delay and ACK options have a dominant impact on the weighted proportionality. Specifically, we develop WCC, a fundamental weighted AIMD-based congestion control building block, which can be implemented via individually modifying AI (WCC-AI) or MD (WCC-MD) behavior. We analyze WCC using extended fluid models, NS3 simulations and Linux kernel implementations with droptail and RED queues, and point out the determinant of performance. Finally, we clarify the influences of dynamic network characteristics on weighted proportionality with sufficient experimental results and summarize a basic law of how to implement weighted AIMD-based congestion control.

While Learning to Rank (LTR) models on top of transformers have been widely adopted to achieve decent performance, it is still challenging to train the model with sufficient data as only an extremely small number of query-webpage pairs could be annotated versus trillions of webpages available online and billions of web search queries everyday. In the meanwhile, industry research communities have released a number of open-source LTR datasets with well annotations but incorporating different designs of LTR features/labels (i.e., heterogeneous domains). In this work, inspired by the recent progress in pre-training transformers for performance advantages, we study the problem of pre-training LTR models using both labeled and unlabeled samples, especially we focus on the use of well-annotated samples in heterogeneous open-source LTR datasets to boost the performance of pre-training. Hereby, we propose S2phere-Semi-Supervised Pre-training with Heterogeneous LTR data strategies for LTR models using both unlabeled and labeled query-webpage pairs across heterogeneous LTR datasets. S2phere consists of a three-step approach: (1) Semi-supervised Feature Extraction Pre-training via Perturbed Contrastive Loss, (2) Cross-domain Ranker Pre-training over Heterogeneous LTR Datasets and (3) End-to-end LTR Fine-tuning via Modular Network Composition. Specifically, given an LTR model composed of a backbone (the feature extractor), a neck (the module to reason the orders) and a head (the predictor of ranking scores), S2phere uses unlabeled/labeled data from the search engine to pre-train the backbone in Step (1) via semi-supervised learning; then Step (2) incorporates multiple open-source heterogeneous LTR datasets to improve pre-training of the neck module as shared parameters of cross-domain learning; and finally, S2phere in Step (3) composes the backbone and neck with a randomly-initialized head into a whole LTR model and fine-tunes the model using search engine data with various learning strategies. Extensive experiments have been done with both offline experiments and online A/B Test on top of Baidu search engine. The comparisons against numbers of baseline algorithms confirmed the advantages of S2phere in producing high-performance LTR models for web-scale search.

Powered by machine learning techniques, online advertising platforms have launched various automated bidding strategy services to facilitate intelligent decision-making for advertisers. However, advertisers experience heterogeneous advertising environments, and thus the unified bidding strategies widely used in both academia and industry suffer from severe unfairness issues, resulting in significant ad performance disparity among advertisers. In this work, to resolve the unfairness issue and improve the overall system performance, we propose a personalized automated bidding framework, namely PerBid, shifting the classical automated bidding strategy with a unified agent to multiple context-aware agents corresponding to different advertiser clusters. Specifically, we first design an ad campaign profiling network to model dynamic advertising environments. By clustering the advertisers with similar profiles and generating context-aware automated bidding agents for each cluster, we can match advertisers with personalized automated bidding strategies. Experiments conducted on the real-world dataset and online A/B test on Alibaba display advertising platform demonstrate the effectiveness of PerBid in improving overall ad performance and guaranteeing fairness among heterogeneous advertisers.

IoT devices are enabled to capture and upload videos with increasing bitrates. Massive IIoT is eager for effective video processing techniques to satisfy the requirements of real-time video services. With the emergence of 5G-unlicensed (5G-U), ultra-low latency video applications become possible. However, existing encoding standards for video services in Web 2.0, such as H.265, are not naturally designed for IIoT video streaming, leading to bandwidth pressure where 5G-U coexists with various other wireless signals. To tackle this problem and to support low-latency video utilization by IIoT video sources, we propose an Adaptive Compression-Reconstruction framework named ACORN, which is based on compressed sensing and recent advances in deep learning. At end nodes, we compress multiple sequential video frames into a single frame to reduce video volume. We design a QoE-aware parameter selection mechanism to deal with volatile network environments during compression. With learnable gated convolution layers and channel-wise soft-thresholding operators, ACORN also builds a real-time reconstruction module. Experimental results reveal that video analytics can be conducted on compressed frames. The reconstruction algorithm in ACORN is with $1-4 \mathrm{~dB}$ improvements. Moreover, both the encoding time cost and the encoded video volume are reduced by more than $4 \times$ under the ACORN framework.

Recent studies show that an end system's traffic may reach a distant anycast site within a global IP anycast system, resulting in high latency. To address this issue, some private and public CDNs have implemented regional IP anycast, a technique that involves dividing content-hosting sites into geographic regions, announcing a unique IP anycast prefix for each region, and utilizing DNS and IP-geolocation to direct clients to CDN sites in their corresponding geographic regions. In this work, we aim to understand how a regional anycast CDN partitions its sites and maps its customers' clients to its sites, and how a regional anycast CDN performs compared to its global anycast counterpart. We study the deployment strategies and the performance of two CDNs (Edgio and Imperva) that currently deploy regional IP anycast. We find that both Edgio and Imperva partition their sites and clients following continent or country borders. Furthermore, we compare the client latency distribution in Imperva's regional anycast CDN with its similar-scale DNS global anycast network, while accounting for and mitigating the relevant deployment differences between the two networks. We find that regional anycast can effectively alleviate the pathology in global IP anycast where BGP routes clients' traffic to distant CDN sites. However, DNS mapping inefficiencies, where DNS returns a sub-optimal regional IP anycast address that does not cover a client's low-latency CDN sites, can harm regional anycast's performance. Finally, we show what performance benefits regional IP anycast can achieve with a latency-based region partition method using the Tangled testbed. When compared to global anycast, regional anycast significantly reduces the 90th percentile client latency by 58.7% to 78.6% for clients across different geographic areas.

In online e-commerce platforms, sponsored ads are always mixed with non-sponsored organic content (recommended items). To guarantee user experience, online platforms always impose strict limitations on the number of ads displayed, becoming the bottleneck for advertising revenue. To boost advertising space, we introduce a novel advertising business paradigm called Augment Advertising, where once a user clicks on a leading ad on the main page, instead of being shown the corresponding products, a collection of mini-detail ads relevant to the clicked ad is displayed. A key component for augment advertising is to design ad auctions to jointly select leading ads on the main page and mini-detail ads on the augment ad page. In this work, we decouple the ad auction into a two-stage auction, including a leading ad auction and a mini-detail ad auction. We design the Potential Generalized Second Price (PGSP) auction with Symmetric Nash Equilibrium (SNE) for leading ads, and adopt GSP auction for mini-detail ads. We have deployed augment advertising on Taobao advertising platform, and conducted extensive offline evaluations and online A/B tests. The evaluation results show that augment advertising could guarantee user experience while improving the ad revenue and the PGSP auction outperforms baselines in terms of revenue and user experience in augment advertising.

Machine learning models have been deployed in mobile networks to deal with massive data from different layers to enable automated network management and intelligence on devices. To overcome high communication cost and severe privacy concerns of centralized machine learning, federated learning (FL) has been proposed to achieve distributed machine learning among networked devices. While the computation and communication limitation has been widely studied, the impact of on-device storage on the performance of FL is still not explored. Without an effective data selection policy to filter the massive streaming data on devices, classical FL can suffer from much longer model training time (4 ×) and significant inference accuracy reduction (7%), observed in our experiments. In this work, we take the first step to consider the online data selection for FL with limited on-device storage. We first define a new data valuation metric for data evaluation and selection in FL with theoretical guarantees for speeding up model convergence and enhancing final model accuracy, simultaneously. We further design ODE, a framework of Online Data sElection for FL, to coordinate networked devices to store valuable data samples. Experimental results on one industrial dataset and three public datasets show the remarkable advantages of ODE over the state-of-the-art approaches. Particularly, on the industrial dataset, ODE achieves as high as 2.5 × speedup of training time and 6% increase in inference accuracy, and is robust to various factors in practical environments.

Approximate query processing (AQP) is one of the key techniques to cope with big data querying problem on account that it obtains approximate answers efficiently. To address non-trivial sample selection and heavy sampling cost issues in AQP, we propose ShadowAQP, an efficient and accurate approach based on attribute-oriented sample size allocation and data generation. We select samples according to group-by and join attributes, and determine the sample size for each group of unique value combinations to improve query accuracy. We design a conditional variational autoencoder model with automatic table data encoding and model update strategies. To further improve accuracy and efficiency, we propose a set of extensions, including parallel multi-round sampling aggregation, data outlier-aware sampling, and dimension reduction optimization. Evaluation results on diversified datasets show that, compared with SOTA approaches, ShadowAQP achieves 5.8× query speed performance improvement on average (up to 12.8×), while reducing query error by 74% on average (up to 95%) at the same time.

Since 2002, hundreds of thousands of vessels have equipped the Automatic Identification System (AIS), which continuously broadcasts its identity and location information for vessel collision avoidance. To utilize these scattered AIS data for further analysis, there are multiple AIS data platforms collecting AIS data from vessels around the world through their satellites and land-based stations. Thus, users can obtain AIS data of vessels from these platforms without dedicated devices. However, existing platforms work in silos, and AIS data is distributed across different platforms, resulting in reduced data availability. In addition, AIS is vulnerable to jamming and spoofing attacks, which can undermine the authenticity of AIS data. In this paper, we propose AISChain, a secure and fast blockchain-based AIS data platform. AISChain adopts consortium blockchain, which only permits those authorized parties (i.e., AIS data providers) to participate in the consensus protocol, and is compatible with current commodity AIS hardware. Since the whole system is co-maintained by multiple authorized parties, AISChain can integrate AIS data resources in a secure way. For avoiding repeated recording of AIS data on the chain, we design the Dynamic Bloom Filter Tree (DBFT) to realize efficient duplication detection in the transaction verification phase. We also propose the dual signature scheme to clarify the AIS data ownership. Moreover, we leverage the geographical location-based blockchain sharding approach to further improve the scalability of AISChain. We implement a prototype of AISChain, and conduct extensive experiments to evaluate the performance of AISChain. Evaluation results show that the search time of DBFT is negligible (4.3 ms) with an extreme low error ratio (0.4%). Meanwhile, AISChain can achieve more than 730 tx/s throughput even when nodes scale to 36. To the best of our knowledge, AISChain is the first work to apply the blockchain technology to secure the AIS data platform.

Recent advances in traffic signal control have witnessed the success of reinforcement learning. However, most of these approaches have focused on vehicle traffic and lack consideration for pedestrians. This can be attributed in part to the fact that the existing underlying technologies are not yet practical to deploy in real-world environments. Vision technologies, for example, can easily be obscured from view in reality. The direction of movement and position of pedestrians is difficult to estimate accurately. The emergence of 6G localization and tracking services offer new opportunities. With this base service, we intend to improve the efficiency, safety, and scalability of multi-intersection traffic signal control with mixed traffic flows. This problem is challenging for its coordination, scalability, and access of new traffic. To solve these challenges, we propose PV-TSC, a distributed reinforcement learning motivated traffic signal control with pedestrian access. We analyze different behaviors of pedestrian traffic, and integrate pedestrian traffic with the proven traffic signal control scheme for vehicle traffic. Finally, we conduct simulation experiments to illustrate the superiority of PV-TSC against classic methods, and further analyze the effectiveness of PV-TSC design by exploring its variants.

Self-supervised graph-level representation learning has recently received considerable attention. Given varied input distributions, jointly learning graphs’ unique and common features is vital to downstream tasks. Inspired by graph contrastive learning (GCL), which targets maximizing the agreement between graph representations from different views, we propose an Ada ptive self-supervised framework, Ada-MIP, considering both M utual I nformation between views (unique features) and inter-graph P roximity (common features). Specifically, Ada-MIP learns graphs’ unique information through a learnable and probably injective augmenter, which can acquire more adaptive views compared to the augmentation strategies applied by existing GCL methods; to learn graphs’ common information, we employ graph kernels to calculate graphs’ proximity and learn graph representations among which the precomputed proximity is preserved. By sharing a global encoder, graphs’ unique and common information can be well integrated into the graph representations learned by Ada-MIP. Ada-MIP is also extendable to semi-supervised scenarios, with our experiments confirming its superior performance in both unsupervised and semi-supervised tasks.

This paper is on designing a compact data structure for multi-set membership testing that allows fast set querying. Multi-set membership testing is a fundamental operation for computing systems. Most existing schemes for multi-set membership testing are built upon Bloom filter and fall short in either storage space cost or query speed. To address this issue, we propose Noisy Bloom Filter (NBF), Error Corrected Noisy Bloom Filter (NBF-E), and Data-driven Noisy Bloom Filter (NBF-D) in this paper. We optimize their misclassification and false positive rates by theoretical analysis and present criteria for selection between NBF, NBF-E, and NBF-D. The key novelty of the three schemes is to store set ID information in a compact but noisy way that allows fast recording and querying and use a denoising method for querying. Especially, NBF-E incorporates asymmetric error-correcting coding techniques into NBF, and NBF-D encodes set ID based on their cardinality. To evaluate NBF, NBF-E, and NBF-D in comparison with the prior art, we conducted experiments using real-world network traces. The results show that NBF, NBF-E, and NBF-D significantly advance the state-of-the-art on multi-set membership testing.

Nowadays, the primary media for information dissemination is shifting to online media. Events usually burst online through multiple modern online media. Therefore, predicting event popularity trends becomes crucial for online platforms to track pubic concerns and make appropriate decisions. However, few researches focus on events popularity prediction from a cross-platform perspective. Challenges origin from vast diversity from events and media, limited access to aligned datasets across different platforms and the great deal of noise in datasets. In this paper, we solve the cross-platform event popularity prediction problem by proposing a model named DancingLines, which is mainly composed of the following three parts. First, we propose TF-SW, a semantic-aware popularity quantification model based on Term Frequency with Semantic Weight, obtaining the event popularity based on Word2Vec and TextRank and generating Event Popularity Time Series(EPTS). Then, we propose DTW-CD, a pairwise time series alignment model derived from DTW with Compound Distance, aligning the EPTS on several platforms. Finally, we aggregate two time series and propose a neural based prediction model implementing Long Short-Term Memory with attention mechanism to obtain accurate predictions. Evaluation results based on large scale real-world datasets demonstrate that DancingLines can efficiently characterize, align, and predict event popularity on cross-platform.

Bloom filter is an efficient data structure for filtering negative keys (keys not in a given set) with substantially small space. However, in real-world applications, there widely exist vulnerable negative keys, which will bring high costs if not being properly filtered, especially when positive keys are added/deleted dynamically. Such problem gets more severe when keys within one set are dynamically added or deleted. Recently, there are works focusing on handling such (vulnerable) negative keys by incorporating learning techniques. These learning-based filters fail to work as the learning techniques can hardly handle incremental insertions or deletions. To address the problem, we proposeSeeSawCountingFilter (SSCF), which is innovated with encapsulating the vulnerable negative keys into a unified counter array named seesaw counter array, and dynamically modulating (or varying) the applied hash functions to guard the encapsulated keys from being misidentified. Moreover, we design ada-SSCF to handle the scenarios where the vulnerable negative keys cannot be obtained in advance. We extensively evaluate our SSCF, which shows that SSCF outperforms the cutting-edge filters by$3\times$on averages regarding accuracy while ensuring a low operation latency. All source codes are in (SSCF-authors).

Due to the increasing demand for scalable and interactive data analytics, column stores have become the de-facto choice in many analytical databases. As a common and fundamental operation in column stores, expression evaluation has a remarkable effect on many queries. To speed up expression evaluation, vectorized techniques such as Single-Instruction-Multiple-Data (SIMD) instructions are widely used. However, there are few works concerning dedicated optimizations for SIMD-based expression evaluation for column stores. In this paper, we propose a runtime optimization framework named ROVEC that enables effective optimizations for SIMD-based expression evaluation. The key idea is to optimize logical expression at execution time, by leveraging lightweight compression and fine-grained statistics associated with the compressed data. ROVEC removes unnecessary type casting and finds the tightest type during evaluation, which maximizes the concurrent operands in SIMD instructions. ROVEC can be applied to many expression-evaluation-intensive operators (e.g., table scan and theta join) for different data types (e.g., numeric, time and string). To validate the effectiveness of ROVEC, we integrate it into a columnar database PolarDB-C. Our evaluation results show that ROVEC improves up to 120% (60% on average) throughput of table scan and up to 50% (30% on average) latency of theta join.

Whilelearning to rank(LTR) has been widely used in web search to prioritize most relevant webpages among the retrieved contents subject to the input queries, the traditional LTR models fail to deliver decent performance due to two main reasons: 1) the lack of well-annotated query-webpage pairs with ranking scores to cover search queries of various popularity, and 2) ill-trained models based on a limited number of training samples with poor generalization performance. To improve the performance of LTR models, tremendous efforts have been done from above two aspects, such as enlarging training sets with pseudo-labels of ranking scores by self-training, or refining the features used for LTR through feature extraction and dimension reduction. Though LTR performance has been marginally increased, we still believe these methods could be further improved in the newly-fashioned “interpolating regime”. Specifically, instead of lowering the number of features used for LTR models, our work proposes to transform original data with random Fourier feature, so as to over-parameterize the downstream LTR models (e.g., GBRank or LightGBM) with features in ultra-high dimensionality and achieve superb generalization performance. Furthermore, rather than self-training with pseudo-labels produced by the same LTR model in a “self-tuned” fashion, the proposed method incorporates the diversity of prediction results between the listwise and pointwise LTR models while co-training both models with a cyclic labeling-prediction pipeline in a “ping-pong” manner. We deploy the proposedCo-trained andOver-parameterizedLTRsystemCOLTRat Baidu search and evaluateCOLTRwith a large number of baseline methods. The results show thatCOLTRcould achieve$\Delta NDCG_{4}$= 3.64%$\sim$4.92%, compared to baselines, under various ratios of labeled samples. We also conduct a 7-day A/B Test using the realistic web traffics of Baidu Search, where we can still observe significant performance improvement around$\Delta NDCG_{4}$= 0.17%$\sim$0.92% in real-world applications.COLTRperforms consistently both in online and offline experiments.

Influence maximization through social networks has aroused tremendous interests nowadays. However, people’s various expressions or feelings about a same idea often cause ambiguity via word of mouth. Consequently, the problem of how to maximize the spread of “effective information” still remains largely open. In this paper, we consider a practical setting where ideas can deviate from their original version to invalid forms during message passing, and make the first attempt to seek a union of users that maximizes the spread of effective influence, which is formulated as an Influence Maximization with Information Variation (IMIV) problem. To this end, we model the information as a vector, and quantify the difference of two arbitrary vectors as a distance by a matching function. We further establish a process where such distance increases with the propagation and ensure the recipient whose vector distance is less than a threshold can be effectively influenced. Due to the NP-hardness of IMIV, we greedily select users that can approximately maximize the estimation of effective propagation. Especially, for networks of small scales, we derive a condition under which all the users can be effectively influenced. Our models and theoretical findings are further consolidated through extensive experiments on real-world datasets.

A crucial privacy-driven issue nowadays is re-identifying anonymized social networks by mapping them to correlated cross-domain auxiliary networks. Prior works are typically based on modeling social networks as random graphs representing users and their relations, and subsequently quantify the quality of mappings through varied cost functions. However, many cost functions are empirically proposed without sufficient theoretical support. For some other works probing the theoretical bound, it remains unknown how to algorithmically meet the demand of such quantifications, i.e., to minimize the cost functions. Besides, only few prior works have discussed the de-anonymization of social networks with communities. We address those concerns in a social network modeling parameterized by community structures that can be leveraged as side information for de-anonymization. Based on the Maximum A Posteriori (MAP) estimation, our first contribution is a series of MAP-based cost functions, which, when minimized, enjoy superiority to previous ones in finding the correct mapping with the highest probability. The feasibility of the cost functions is then for the first time algorithmically characterized. We prove the general multiplicative inapproximability and thus propose two heuristics, which, respectively, enjoy an$\epsilon$-additive approximation and a conditional optimality in carrying out successful user re-identification. Our theoretical findings are also empirically validated under classical synthetic and real-wrold social networks. Both theoretical and empirical observations manifest the importance of community in enhancing privacy inferencing.

Several recent works have studied mobile charging under the “one-to-many” charging pattern where a single charger can charge multiple devices simultaneously. However, most of them focus on path planning and charging time allocation, but overlook the underlying dependence of the charging efficiency on initial deployment positions of chargers. This paper studies the problem ofPlacing directional wIreless chargers withLimited mObiliTy (PILOT), that is, maximize the overall charging utility for a set of static rechargeable devices on a 2D plane by determining deployment positions, stop positions and orientations, and portions of time for all deployed chargers that can move in a limited area after their deployment. To the best of our knowledge, we are the first to study placement of mobile directional chargers under the “one-to-many” pattern. To address PILOT, we propose a$(\frac{1}{2}-\epsilon)$-approximation algorithm. First, we present a method to approximate nonlinear charging power of chargers, and further propose an approach to construct Maximal Covered Set uniform subareas to reduce the infinite continuous search space for stop positions and orientations to a finite discrete one. Second, we present geometrical techniques to further reduce the infinite solution space for candidate deployment positions to a finite one without performance loss, and transform PILOT to a mixed integer nonlinear programming problem. Finally, we propose a linear programming based greedy algorithm to address it. Simulation and experimental results show that our algorithm outperforms six comparison algorithms by$19.74 \% \sim 500.01 \%$.

More and more IoT data is being traded online in cloud-based data marketplaces due to the fast-growing market demand. Within the current data selling mechanisms, data consumers have difficulties in making purchasing decisions due to uncertain IoT data quality and inflexible pricing interface. To resolve these issues, potential solutions could be to launch data demonstrations and release free sampling data to reduce the uncertainty about data quality, and to charge based on the volume of data actually used to enable flexible pricing. However, there is still no clear understanding of economic benefits of these mechanisms. In this paper, we design the optimal data selling mechanisms for IoT data exchange, and derive the following two results. First, whether to deploy a data demonstration and how much free sampling data to release depend on the extent of data consumers' inaccuracy perceptions for data quality, which varies over a wide range in IoT applications. We found that the data vendor has no incentive to conduct these strategies if data consumers extremely overestimate data quality. Second, although flexible data pricing mechanisms provide convenience for real-time and streaming IoT data exchange, it brings less economic benefits to the data vendor compared with the fixed pricing scheme, which sells the whole data set with a fixed price. We evaluate the optimal selling mechanisms on a real-world Taxi GPS data set, and evaluation results verify the insights derived from our theoretical analysis.

With the increasing popularity of mobile devices, each user is able to conveniently acquire messages from others, and share diverse forms of information, like texts, images, or videos through online mobile apps. The full freedom of speech makes a great amount of truth (i.e., true information) and rumor (i.e., false information) propagate rapidly in a hybrid way through mobile platforms. As a huge variety of information floods pouring over us each day, identifying the authenticity of massive events becomes a necessary task to maintain the stability of Mobile Social Networks (MSNs). An important way to realize it is to trace their diffusions and make judgements according to the reliability of sources. With this regard, this paper proposes a diffusion model that characterizes the simultaneous diffusion of both truth and rumor in realistic MSNs, and makes the first attempt to figure out their respective sources. The problem of interest can be stated as: Given an outcome of cascade of both truth and rumor in MSNs, i.e., a set of nodes that might be the ignorant, the spreader of truth or rumor, or simply the silent receiver, how can we infer both truth sources and rumor sources? Different from previous sources detection works considering single type of nodes, the interplay between truth diffusions and rumor diffusions makes the conventional methods not work. To answer this question, we aim to maximize thesimilarity index, i.e., the number of nodes possessing the same states between the resulting network triggered by our estimated sources with the proposed diffusion model and the given observation network. Compared with existing techniques to trace diffusions of truth or rumor, it is much harder to find two kinds of sets at the same time, including truth sources and rumor sources, due to two primary reasons: (i) our biset optimization makes the submodularity techniques fail; (ii) our objective function is proven to be non-bisubmodular. To overcome above limitations, we first convert the objectivesimilarity indexinto a bisubmodular function by virtue of set covering. Based on this, we propose an approximation algorithm called Truth and Rumor Sources Detection (TRSD) algorithm via multiple reverse samplings with a provable$\frac{1}{4(1+\epsilon)^2}$approximation ratio. Further, a novel “time reversal” sources optimization strategy is proposed to converge the number of output sources from TRSD to a steady state. The effectiveness of our models and algorithms are empirical validated in two various datasets, from which we observe an up to 15% ofsimilarity indexgain as well as a narrowed down gap 0.6% to the ground truth.

As a low-power and low-cost wireless protocol, the promising ZigBee has been widely used in sensor networks and cyber-physical systems. Since ZigBee based networks usually adopt tree or cluster topology, the convergecast scenarios are common in which multiple transmitters send packets to one receiver, leading to the severe collision problem. The conventional ZigBee adopts carrier sense multiple access with collisions avoidance to avoid collisions, which introduces additional time/energy overhead. The state-of-the-art methods resolve collisions instead of avoidance, in which mZig decomposes a collision by the collision itself and reZig decodes a collision by comparing with reference waveforms. However, mZig falls into high decoding errors only exploiting the signal amplitudes while reZig incurs high computational complexity for waveform comparison. In this paper, we propose CmZig to embrace channel estimation in multiple-packet reception (MPR) of ZigBee, which effectively improves MPR via lightweight computing used for channel estimation and collision decomposition. First, CmZig enables accurate collision decomposition with low computational complexity, which uses the estimated channel parameters modeling both signal amplitudes and phases. Second, CmZig adopts reference waveform comparison only for collisions without chip-level time offsets, instead of the complex machine learning based method. We implement CmZig on USRP-N210 and establish a six-node testbed. Results show that CmZig achieves a bit error rate in the order of$10^{-3}$and more than 80% packet reception rate, which outperforms state-of-the-art mZig.

The omnipresence of information cascading process in mobile social networking applications makes the identification of a small set$S$of influential users, which is widely believed to trigger the information outbreak, always an crucial issue in various applications such as the mobile advertising and viral marketing. Formulated as Influence maximization (IM) in 2003, this NP-hard problem has received a multitude of studies with diverse angles. However, these works often unable to provide reliable solutions, due to the loss of an exact metric for evaluating users’ contributions on information cascading in the state-of-the-art sampling based IM schemes. In this paper, we evaluate users in IM based on the collective influence (CI), a metric on the structural features of the users in network graph that reflects the contributions of the users’ neighborhoods on shaping collective dynamics of the users over the whole network. For conducting the influencer identification under probabilistic diffusion model based on the CI, we specify a quantified structural feature of the most influential users from the scope of diffusion over the whole network, and reveal that the structural influence power (CI value) of each user is a weighted cumulation of the diffusion probabilities from neighbors within certain hops. Utilizing CI, we design a novel algorithm which identifies the influencers via iteratively choosing the users with top CI values. Moreover, we point out that directly computing CI values requires to traverse the network which is originally represented by a high-dimensional matrix, and leads to huge complexity of influencer identification. To improve scalability, we further trade precision for efficiency by incorporating network embedding, a dimensionality reduction technology for networks, into algorithm design, and propose a minor variant, where CI is jointly recapitulated by low-dimensional user representations and user degrees. The superiority of our algorithms is empirically validated over 8 datasets, with an increment in influence size up to 50 percent and a comparable or even less running time comparing with existing baselines.

A major challenge confronting today’s distributed metadata management schemes is how to meet the dynamic requirements of various applications through effectively mapping and migrating metadata nodes to different metadata servers (MDS’s). Most of the existing works dynamically reallocate nodes to different servers adopting history-based coarse-grained methods, failing to make a timely and efficient update on the distribution of nodes. In this paper, we present the first deep reinforcement learning-leveraged distributed metadata management scheme, AdaM, to address the aforementioned dilemma. AdaM is an adaptive fine-grained metadata management scheme that trains an actor-critic network to migrate “hot” metadata nodes to different MDS’s based on its observations of the current “states” (i.e., access pattern, the structure of namespace tree and current distribution of nodes on MDS’s). Adaptive to varying access patterns, AdaM can automatically migrate hot metadata nodes among servers to keep load balancing while maintaining metadata locality. Besides, we propose a self-adaptive metadata cache policy, which dynamically combines the two strategies of managing caches on the server side and the client side to gain better query performance. Last but not least, we design a distributed metadata processing 2PC Protocol called MST-based 2PC to ensure data consistency. Experiments on a real-world dataset demonstrate the superiority of our proposed method over other schemes.

A controller needs to solve the multi-commodity flow problem and globally update the network under tight time constraints to maintain optimal network configurations. This centralized optimization in data centers involves many variables and constraints, which has a slow convergence speed and little scalability. In this paper, we propose MiFi, which aims to Minimize Flow cost or intuitively average transmission delay (delay or latency of flows), under reconfiguration budget constraints in data centers. Thus, we formulate this optimization problem as a constrained Markov Decision Process and propose a set of algorithms to solve it in a scalable manner. We first develop a propagation algorithm to identify the flows mostly affected in terms of latency and configuration in the next update. Then, we set a limitation range (the subset of switches requiring network updates) for updating them to improve adaptability and scalability by updating a less number of flows each time to achieve fast operations. Further, based on the Drift-Plus-Penalty method in Lyapunov theory, we propose a heuristic policy without prior information of flow demand and a renewal policy with a performance guarantee to minimize the additive optimality gap. To the best of our knowledge, MiFi is the first paper that studies the range and frequency of flow reconfigurations, which has both theoretical and practical significance in the area. Emulations and numerical simulations, which are much better than the estimated theoretical bound, show that MiFi outperforms the state of the art algorithms in terms of latency by over 45% while making improvements in adaptability and scalability.

Virtualized network functions (VNFs) enable software applications to replace traditional middleboxes, which are more flexible and scalable in the network service provision. This paper focuses on ensuring Service Availability via Failure Elimination (SAFE) using VNF scaling, that is, given the resource requirements of VNF instances, finding an optimal and robust instance consolidation strategy, which can recover from one instance failure quickly. To address the above problem, we present a framework based on rounding and dynamic programming. First, we discretize the range of resource requirements into several sub-ranges, and thus the number of instance types becomes a constant. Second, we further reduce the number of instance types by gathering several small instances into a bigger one. Third, we propose an algorithm built on dynamic programming to solve the instance consolidation problem with a limited number of instance types. Finally, we set up a testbed to profile the functional relationship between the resource and the throughput for different types of VNFs, and conduct simulations to validate our theoretical results according to profiling results. The simulation results show that our algorithm outperforms the standby deployment model by 27.33% on average in terms of the number of servers required. Furthermore, SAFE has marginal overheads, around 7.22%, compared to the instance consolidation strategy without VNF backup consideration.

Wireless home surveillance cameras are gaining popularity in elderly/baby care and burglary detection in an effort to make our life safer than ever before. However, even though the camera’s traffic is encrypted, attackers can still infer what the residents are doing at home. Although this security loophole has been reported, it does not attract much attention from the public, as it requires the attacker to be in close proximity to the camera and have some prior knowledge about the victims. Due to these requirements, the attacker has a low chance of success in the real world. In this article, we argue that the capability of attackers has been greatly underestimated. First, the attacker can leverage the characteristics of video transport protocols to recover the metadata of missing packets. Second, the attacker can build the inference model using the public datasets and adapt the model to the real traffic. Thus, the attacker can launch the attack at a distance from the camera, without prior knowledge about the victim. We also implement this attack scenario and verify that the attacker can infer the victims’ activities at a distance as large as 40 m without any knowledge about the victim, neither personal nor environmental.

Continuous domestic respiration monitoring provides vital information for diagnosing assorted diseases. In this article, we introduce RespTracker , the first continuous, multiple-person respiration tracking system in domestic settings using acoustic-based COTS devices. RespTracker uses a multi-stage algorithm to separate and recombine respiration signals from multiple paths so that it can track the respiration rate of multiple moving subjects. And it leverages features from multiple dimensions to separate different users in the same area. Our experimental results show that our two-stage algorithm can distinguish the respiration of at least four subjects and cover a distance of three meters.

Remote Direct Memory Access (RDMA) has been widely deployed in datacenters for its high performance. Large-scale high performance cloud services built on geographically distributed datacenters require long-range RDMA for performance requirements. However, existing RDMA solutions can hardly satisfy the stringent requirements of the emerging large-scale high-performance cloud services built on geo-distributed datacenters in terms of throughput and delay. On the one hand, lossless RDMA suffers from a deep buffer and potential suboptimal throughput for inter-datacenter traffic due to delayed response to Priority Flow Control (PFC) messages. On the other hand, lossy RDMA with selective retransmissions suffers from poor performance when multiple flows with different round-trip times (RTTs) coexist in cross-datacenter scenarios. This article proposesSwing, which expands the high-performance lossless RDMA to long-distance links through PFC-Relay.Swingensures the throughput of long-distance links while minimizing the buffer requirement for long-range RDMA. It enables long-range RDMA without making any modifications to existing in-datacenter networks. The evaluation shows thatSwingcan reduce the average flow completion time (FCT) by 14%-66% in a variety of traffic scenarios.

Nowdays, it is prevalent to train deep learning models in cloud-native platforms that actively leverage containerization and orchestration technologies for high elasticity, low and flexible operation cost, and many other benefits. However, it also faces new challenges and our work is focusing on those related to I/O throughput for training, including complex data access, lack of matching dynamic I/O requirement, and inefficient I/O resource scheduling across different jobs. We proposeFluid, a cloud-native platform that provides DL training jobs with high-level data abstraction calledFluid Datasetto access training data from heterogeneous sources with elastic data acceleration. In addition, it comes with an on-the-fly cache system autoscaler that can match the online training speed and increase the number of cache replicas adaptively to alleviate I/O bottlenecks. To improve the overall performance of multiple DL jobs, Fluid co-orchestrate the data cache and DL jobs by arranging job scheduling in an appropriate order and can also schedule data cache and DL jobs on the same node to realize cache affinity. Experimental results show significant performance improvement of each individual DL job which uses dynamic computing resources with Fluid. For scheduling multiple DL jobs with same datasets, Fluid achieves around 2x performance speedup when integrated with existing widely-used and cutting-edge scheduling solutions through the appropriate job scheduling order. Besides, the cache affinity scheduling policy also improves job execution performance significantly. Fluid is now an open source project hosted by Cloud Native Computing Foundation (CNCF) with many production adopters.

The exponential growth of Massive Open Online Courses (MOOCs) surges the needs of advanced models for personalized Online Education Services (OES). Existing solutions successfully recommend MOOCs courses via deep learning models, they however generate weak “course embeddings” with original profiles, which contain noisy and few enrolled courses. On the other hand, existing algorithms provide recommendation orders according to the score of each course while ignoring personalized demands of users. To tackle the above challenges, we propose aMetaHierarchicalReinforcedRankingapproachMHRR, which consists of a meta hierarchical reinforcement learning pre-trained mechanism and an over-parameterized ranking regressor to enhance the representation learning of courses and learners while refining the ranking result of recommended courses. Specifically,MHRRcombines a user profile reviser and a meta embedding generator to provide course embedding representation enhancement for recommender services. Furthermore,MHRRtransforms learned representations generated from recommender services with Gaussian kernel approximation to over-parameterize the downstream learning to rank (LTR) models with representations in ultra-high dimensionality. We deployMHRRon a real-world MOOCs platform and evaluate it with a large number of baseline models. The results show thatMHRRoutperforms baseline algorithms on two major metrics, including Hit Ratio (HR) and Normalized Discounted Cumulative Gain (NDCG). Also, we conduct a 7-day online evaluation using the realistic traffic of a large-scale real-world MOOCs platform, where we can still observe significant improvement in real-world applications.MHRRperforms consistently both in the online and offline evaluation.

Multi-task learning (MTL) has been widely used in recommender systems, wherein predicting each type of user feedback on items (e.g, click, purchase) are treated as individual tasks and jointly trained with a unified model. Our key observation is that the prediction results of each task may contain task-specific knowledge about user’s fine-grained preference towards items. While such knowledge could be transferred to benefit other tasks, it is being overlooked under the current MTL paradigm. This paper, instead, proposes a Cross-Task Knowledge Distillation framework that attempts to leverage prediction results of one task as supervised signals to teach another task. However, integrating MTL and KD in a proper manner is non-trivial due to several challenges including task conflicts, inconsistent magnitude and requirement of synchronous optimization. As countermeasures, we 1) introduce auxiliary tasks with quadruplet loss functions to capture cross-task fine-grained ranking information and avoid task conflicts, 2) design a calibrated distillation approach to align and distill knowledge from auxiliary tasks, and 3) propose a novel error correction mechanism to enable and facilitate synchronous training of teacher and student models. Comprehensive experiments are conducted to verify the effectiveness of our framework in real-world datasets.

A urine bag is a common medical appliance widely used in clinical treatment for patients with urinary frequency, urgency, or incontinence problems. The real-time monitoring and alarming of the liquid volume in the bag are necessary to avoid the infection caused by urine reflux. Current monitoring relies on manual observations of doctors and medical attendants, but it has problems with timeliness, especially in surgery or at night. In this paper, we propose Wi-Uro, a real-time, low-cost, and accurate urine bag monitoring and alarming system with commodity Wi-Fi devices. To solve the random deformation of urine bags possibly due to human postures, it integrates Wi-Fi Channel State Information (CSI) with tension sensor readings to infer liquid volume in bags. Once the volume is exceeding the threshold, Wi-Uro system will prompt the alarm information immediately. Evaluations based on real urine bags show a satisfactory estimation accuracy (over 95%), which means that Wi-Uro can be applied in routine patient care.

In the last decade, MIMO spatial multiplexing and distributed beamforming play a significant role in improving data throughput through cooperative transmission. It has been widely used in wireless communication, especially in 6G. However, the distributed uplink beamforming is still an open problem in highly dynamic environments. However, the proposed 6G technology represents the further integration of deep learning and wireless communication. In this paper, we propose Argute Distributed Uplink Beamforming (ArguteDUB), which uses a feedback algorithm with an offline-trained deep learning model to implement highly dynamic distributed uplink beamforming for the Internet of Vehicles (IoV) in 6G. Specifically, each vehicle enables the base station (BS)/access point (AP) to separate different channel state information (CSI) by inserting orthogonal sequences into the sending data. The BS adopts deep learning to filter the noise and predict the beamforming weight to achieve phase synchronization. Unlike traditional distributed uplink beamforming, ArguteDUB can be adapted to the highly dynamic time-varying channels. The simple network structure ensures the fast response of ArguteDUB. In addition, we make ArguteDUB Orthogonal Frequency Division Multiplexing (OFDM) compatible so that it can be easily deployed in 6G networks. Our evaluation shows that ArguteDUB has an SNR gain of about 5dB to 5.3dB over the single vehicle transmission mode.

Nowdays, it is prevalent to train deep learning (DL) models in cloud-native platforms that actively leverage containerization and orchestration technologies for high elasticity, low and flexible operation cost, and many other benefits. However, it also faces new challenges and our work is focusing on those related to I/O throughput for training, including complex data access with complicated performance tuning, lack of cache capacity with specialized hardware to match its high and dynamic I/O requirement, and inefficient I/O resource sharing across different training jobs. We propose Fluid, a cloud-native platform that provides DL training jobs with a data abstraction called Fluid Dataset to access training data from heterogeneous sources in a unified manner with transparent and elastic data acceleration powered by auto-tuned cache runtimes. In addition, it comes with an on-the-fly cache system autoscaler that can intelligently scale up and down the cache capacity to match the online training speed of each individual DL job. To improve the overall performance of multiple DL jobs, Fluid can co-orchestrate the data cache and DL jobs by arranging job scheduling in an appropriate order. Our experimental results show significant performance improvement of each individual DL job which uses dynamic computing resources with Fluid. In addition, for scheduling multiple DL jobs with same datasets, Fluid gives around 2x performance speedup when integrated with existing widely-used and cutting-edge scheduling solutions. Fluid is now an open source project hosted by Cloud Native Computing Foundation (CNCF) with adopters in production including Alibaba Cloud, Tencent Cloud, Weibo.com, China Telecom, etc.

The approximate membership query (AMQ) data structure is a kind of space-efficient probabilistic data structure. It can approximately indicate whether an element exists in a set. The AMQ data structure has been widely used in database indexing, network security, IoT applications, etc. Resizing is an extensively utilized operation of the AMQ data structure, but it can lead to system performance degradation. We summarize two main problems that lead to such degradation. Specifically, one of them is that the resizing operation can block other operations, while the other is that the performance of AMQ structures will deteriorate after multiple resizing operations. However, existing related work cannot alleviate both of them. Therefore, we propose a novel AMQ data structure called bamboo filter, which can alleviate the two problems simultaneously. Bamboo filters can insert, search and delete an element in constant time. Moreover, bamboo filters can dynamically resize in a fine-grained way according to the number of contained elements. Experimental results show that bamboo filters significantly outperform state-of-the-art resizable AMQ data structures in insertion, lookup, and deletion operations. For example, bamboo filters achieve$\mathbf{2.46}\times$lookup throughput of the dynamic cuckoo filter, on average.

In Click-through rate (CTR) prediction models, a user’s interest is usually represented as a fixed-length vector based on her history behaviors. Recently, several methods are proposed to learn an attentive weight for each user behavior and conduct weighted sum pooling. However, these methods only manually select several fields from the target item side as the query to interact with the behaviors, neglecting the other target item fields, as well as user and context fields. Directly including all these fields in the attention may introduce noise and deteriorate the performance. In this paper, we propose a novel model named AutoAttention, which includes all item/user/context side fields as the query, and assigns a learnable weight for each field pair between behavior fields and query fields. Pruning on these field pairs via these learnable weights lead to automatic field pair selection, so as to identify and remove noisy field pairs. Though including more fields, the computation cost of AutoAttention is still low due to using a simple attention function and field pair selection. Extensive experiments on the public dataset and Tencent’s production dataset demonstrate the effectiveness of the proposed approach.

Mobile edge computing (MEC) is emerging as a novel computing paradigm that pushes network resources (such as computation and storage resources) away from the centralized data center to distributed edge servers. By hiring various resources of nearby edge servers, the MEC provides high-bandwidth and low-latency network services for mobile users. As numerous mobile users may compete for limited edge servers’ resources, to improve the resource utilization of the MEC system, it is very critical to investigate an effective user allocation policy. Previous studies mainly focus on investigating offline user allocation policies. However, mobile users may arrive online, and the MEC should be able to allocate these users online too. In a real-world MEC environment, online allocation decisions should not be made entirely in the dark. The historical user requests which may contain powerful hints about future user requests, can be adopted to assist in making allocation decisions. In this paper, we take the historical data into account and study the history-assisted online user allocation strategy. Specifically, we formulate the user allocation problem with a comprehensive model and show its hardness. Then, we present an online algorithm named HOUA to allocate mobile users according to both the online arrived user requests and the historical user requests. The competitive ratio of HOUA is proved. To further verify the effectiveness of HOUA, we conduct experiments on a widely-used real-world dataset. We show that HOUA can allocate more mobile users and achieve high resource rental revenue compared with the other approaches.

Nowadays, Database-as-a-Service (DBaaS) has become more and more popular among users as it can largely reduce the complexity of managing databases and applications. Considering the increasing complexity of different applications, system management and automation such as self-provisioning and performance tuning can be challenging. To better achieve autonomous optimization of the system, the ability to predict future workload patterns is of great essence. In this paper, we propose a novel lightweight probabilistic workload forecasting framework (LIGHTPRO) that is easy to train and robust, to help the system predict future workload patterns, leveraging multi-head attention mechanism and convolution operations. Experiments on real-world query traces demonstrate the superiority of LIGHTPRO in reducing training time and capturing both long-term and short-term temporal patterns of the workload compared with other baselines.

Collaborative filtering (CF), as a standard method for recommendation with implicit feedback, tackles a semi-supervised learning problem where most interaction data are unobserved. Such a nature makes existing approaches highly rely on mining negatives for providing correct training signals. However, mining proper negatives is not a free lunch, encountering with a tricky trade-off between mining informative hard negatives and avoiding false ones. We devise a new approach named as Hardness-Aware Debiased Contrastive Collaborative Filtering (HDCCF) to resolve the dilemma. It could sufficiently explore hard negatives from two-fold aspects: 1) adaptively sharpening the gradients of harder instances through a set-wise objective, and 2) implicitly leveraging item/user frequency information with a new sampling strategy. To circumvent false negatives, we develop a principled approach to improve the reliability of negative instances and prove that the objective is an unbiased estimation of sampling from the true negative distribution. Extensive experiments demonstrate the superiority of the proposed model over existing CF models and hard negative mining methods.

This paper studies the problem of connectivity maintenance in adversarial uncertain networks, where a defender prevents the largest connected component from being decomposed by an attacker. In contrast with its deterministic counterpart, connectivity maintenance in an uncertain network involves additional testing on edges to determine their existence. To this end, by modeling a general uncertain network as a random graph with each edge associated with an existence probability and a testing cost, our goal is to design a general adaptive defensive strategy to maximize the expected size of the largest remaining connected component with minimum expected testing cost and, moreover, the strategy should be independent of the attacking patterns. The computational complexity of the connectivity maintenance problem is unraveled by proving its NP-hardness. To accurately tackle the problem, based on dynamic programming we first propose an optimal defensive strategy for a specific class of uncertain networks with uniform testing costs. Thereafter multi-objective optimization is adopted to generalize the optimal strategy for general uncertain networks through weighted sum of normalized size and cost. Due to the prohibitive price of an optimal strategy, two approximate defensive strategies are further designed to pursue decent performance with quasilinear complexity. We first derive a heuristic approach by quantifying the edge vulnerability through an analogy from the degree centrality in deterministic networks to the probability degree and connectivity weight in uncertain networks. For performance guarantee, we then devise an adaptive greedy policy incorporating the minimax rule from game theory, which minimizes the possible loss suffered by the defender in a worst-case scenario caused by the attacker and has an approximation ratio of (1 − 1/e). Extensive experiments on both synthetic and real-world network datasets under diverse attacking patterns demonstrate the superiority of the proposed strategies over baselines.

Mobile health (mHealth) applications, benefiting from mobile computing, have emerged rapidly in recent years, and generated a large volume of mHealth data. However, these valuable data are dispersed across isolated devices or organizations, which hinders discovering insights underlying the aggregated data. Considering the online characteristics of mHealth tasks, there is an urgent need for online data acquisition. In this paper, we present the first online data Valuation And Pricing mechanism, namely VAP, to incentive users to contribute mHealth data for machine learning (ML) tasks in mHealth systems. Under the framework of Bayesian ML, we propose a new metric based on the concept of entropy, to evaluate data valuation during model training in an online manner. In proportion to the data valuation, we then determine payments as compensations for users to contribute their data. We formulate this pricing problem as a contextual multi-armed bandit with the goal of profit maximization and propose a new algorithm based on the characteristics of pricing. We also extend VAP to general ML models. Finally, we have evaluated VAP on two real-world mHealth data sets. Evaluation results show that VAP outperforms the state-of-the-art valuation and pricing mechanisms in terms of computational complexity and extracted profit.

Wireless sensor networks and mobile crowdsensing are two important paradigms in urban dynamic sensing. In both sensing paradigms, task allocation is a significant problem that may affect the completion quality of sensing tasks. In this paper, we focus on allocating sensing tasks for mobile crowdsensing. We propose a system model to illustrate how the allocation mechanism works in crowdsensing and define the task allocation problem as maximum profit routing for mobile crowdsensing (MPRMC). The NP-hardness of MPRMC is proved based on a special case analysis. To solve MPRMC, we consider the problem in two special cases: Temporal-MPRMC and Spatial-MPRMC. Correspondingly, we propose a fully polynomial-time approximation scheme for Temporal-MPRMC and a 2Ɣ-approximation for Spatial-MPRMC, respectively. Then, we generalize the special cases and give a 2Ɣ lg2n-approximation for MPRMC. To the best of our knowledge, our algorithms have the best approximation ratios for MPRMC. Finally, we conduct various experiments to validate the effectiveness of our designs.

Cloud-based learning is currently the mainstream in both academia and industry. However, the global data distribution, as a mixture of all the users' data distributions, for training a global model may deviate from each user's local distribution for inference, making the global model non-optimal for each individual user. To mitigate distribution discrepancy, on-device training over local data for model personalization is a potential solution, but suffers from serious overfitting. In this work, we propose a new device-cloud collaborative learning framework under the paradigm of domain adaption, called MPDA, to break the dilemmas of purely cloud-based learning and on-device training. From the perspective of a certain user, the general idea of MPDA is to retrieve some similar data from the cloud's global pool, which functions as large-scale source domains, to augment the user's local data as the target domain. The key principle of choosing which outside data depends on whether the model trained over these data can generalize well over the local data. We theoretically analyze that MPDA can reduce distribution discrepancy and overfitting risk. We also extensively evaluate over the public MovieLens 20M and Amazon Electronics datasets, as well as an industrial dataset collected from Mobile Taobao over a period of 30 days. We finally build a device-tunnel-cloud system pipeline, deploy MPDA in the icon area of Mobile Taobao for click-through rate prediction, and conduct online A/B testing. Both offline and online results demonstrate that MPDA outperforms the baselines of cloud-based learning and on-device training only over local data, from multiple offline and online metrics.

We focus on federated learning in practical recommender systems and natural language processing scenarios. The global model for federated optimization typically contains a large and sparse embedding layer, while each client’s local data tend to interact with part of features, updating only a small submodel with the feature-related embedding vectors. We identify a new and important issue that distinct data features normally involve different numbers of clients, generating the differentiation of hot and cold features. We further reveal that the classical federated averaging algorithm (FedAvg) or its variants, which randomly selects clients to participate and uniformly averages their submodel updates, will be severely slowed down, because different parameters of the global model are optimized at different speeds. More specifically, the model parameters related to hot (resp., cold) features will be updated quickly (resp., slowly). We thus propose federated submodel averaging (FedSubAvg), which introduces the number of feature-related clients as the metric of feature heat to correct the aggregation of submodel updates. We prove that due to the dispersion of feature heat, the global objective is ill-conditioned, and FedSubAvg works as a suitable diagonal preconditioner. We also rigorously analyze FedSubAvg’s convergence rate to stationary points. We finally evaluate FedSubAvg over several public and industrial datasets. The evaluation results demonstrate that FedSubAvg significantly outperforms FedAvg and its variants.

Spectrum sensing plays a crucial role in spectrum monitoring and management. However, due to the expensive cost of high-speed ADCs, wideband spectrum sensing is a long-standing challenge. In this paper, we present how to transform Ultra-wideband (UWB) devices into a spectrum sensor which can provide wideband spectrum monitoring at a low cost. Compared with the expensive high-speed ADCs which cost at least hundreds of dollars, a UWB device is only several dollars. As the low-cost UWB technology is not originally designed for spectrum sensing, we address the inherent limitations of low-cost devices such as limited memory, low SPI speed and low accuracy, and show how to obtain spectrum occupancy information from the noisy and spurious UWB channel impulse response. In this paper, we present WISE, which not only can give accurate channel occupancy information, but also can precisely estimate the signal power and bandwidth. WISE can also detect fleeting radar signals. We implement WISE and perform extensive evaluations with both controlled experiments and field tests. Results show that WISE can sense up to 900MHz bandwidth and the power estimation error is less than 3dB. WISE can also accurately detect busy 5G channels. We believe that WISE provides a new paradigm for low-cost wideband spectrum sensing, which is critical for large-scale fine-grained spectrum monitoring.

Spectrum sensing plays a crucial role in spectrum monitoring and management. However, Due to the expensive cost of high-speed ADCs, wideband spectrum sensing is a long-standing challenge. In this demo, we present how to transform the low-cost Ultrawideband (UWB) devices into a spectrum sensor and showcase WISE, a low-cost wideband spectrum sensing system, which not only can give accurate channel occupancy information, but also can precisely estimate the signal power and bandwidth. Our demo will show that WISE can sense up to 900MHz bandwidth and the power estimation error is less than 3dB. WISE can also accurately detect busy 5G channels and fleeting radar signals. We believe that WISE provides a new paradigm for low-cost wideband spectrum sensing, which is critical for large-scale fine-grained spectrum monitoring.

Network measurement is important to data center operators. Most existing efforts focus on developing new implementation schemes for measurement tasks. Little attention is paid to on-the-fly task reconfiguration. Due to resource constraints, it is impossible to configure all needed tasks at start-up and dynamically turn on/of them. To support real-time reconfiguration of many different tasks, a key observation is that it is unnecessary to bind a task and its implementation at the compilation phase. We design FlyMon, the first sketch-based measurement system that can make on-the-fly reconfigurations on a large set of measurement tasks. FlyMon introduces the concept of Composable Measurement Units (CMUs), which are general operation units that support reconfigurable implementation for measurement tasks combined from different flow keys and flow attributes. FlyMon maps the design of CMUs to programmable switches' data planes so that the number of compacted CMUs can be maximized. FlyMon also provides dynamic memory management. We prototype FlyMon on Tofino and currently enable four frequently used flow attributes. Each CMU Group (with 3 CMUs) can concurrently perform up to 96 isolated measurement tasks with less than 8.3% hardware resources. The tasks can be deployed with configurable memory size at the millisecond level. By cross-stacking, FlyMon can deploy up to 27 CMUs in one pipeline of Tofino.

Click-through rate (CTR) prediction plays an important role in online advertising and recommendation systems, which aims at estimating the probability of a user clicking on a specific item. Feature interaction modeling and user interest modeling methods are two popular domains in CTR prediction, and they have been studied extensively in recent years. However, these methods still suffer from two limitations. First, traditional methods regard item attributes as ID features, while neglecting structure information and relation dependencies among attributes. Second, when mining user interests from user-item interactions, current models ignore user intents and item intents for different attributes, which lacks interpretability. Based on this observation, in this paper, we propose a novel approach Hierarchical Intention Embedding Network (HIEN), which considers dependencies of attributes based on bottom-up tree aggregation in the constructed attribute graph. HIEN also captures user intents for different item attributes as well as item intents based on our proposed hierarchical attention mechanism. Extensive experiments on both public and production datasets show that the proposed model significantly outperforms the state-of-the-art methods. In addition, HIEN can be applied as an input module to state-of-the-art CTR prediction methods, bringing further performance lift for these existing models that might already be intensively used in real systems.

The Narrow Band-Internet of Things (NB-IoT) is a wireless protocol with low-power, low-cost, and wide coverage, which has a broad application prospect in fields like smart city or manufacturing. In order to expand the network coverage, the repetition mechanism is designed in NB-IoT, where the same information is required to be transmitted repeatedly for the accurate decoding of received signals in low signal-to-noise ratio (SNR) environments. However, such repetition mechanism will cause a dramatic decrease in the spectrum utilization, which will be greatly reduced by up to 1/128. Literature has explored efficient coding methods, such as Non-Orthogonal Multiple Access (NOMA), but they can only achieve 55% maximum improvement on spectral efficiency.To further increase the spectrum utilization, this paper propose a novel uplink repetition mechanism design for NB-IoT, namely mNB-IoT. It can enable multiple user equipments (UEs) to transmit data to base stations in parallel using the same subcarriers. Based on the timing synchronization of NB-IoT, the original repetition mechanism is changed to the repetition based on orthogonal sequences to realize concurrent transmission among multiple UEs. The repetition intervals and times are then unified among UEs to maximize spectral efficiency. Experiments built upon 3 USRP devices and simulations on 128 UEs have proven the efficiency of mNB-IoT in realizing the repetition mechanism without spectral efficiency reduction.

In online advertising, auto-bidding has become an essential tool for advertisers to optimize their preferred ad performance metrics by simply expressing high-level campaign objectives and constraints. Previous works designed auto-bidding tools from the view of single-agent, without modeling the mutual influence between agents. In this paper, we instead consider this problem from a distributed multi-agent perspective, and propose a general \underlineM ulti-\underlineA gent reinforcement learning framework for \underlineA uto-\underlineB idding, namely MAAB, to learn the auto-bidding strategies. First, we investigate the competition and cooperation relation among auto-bidding agents, and propose a temperature-regularized credit assignment to establish a mixed cooperative-competitive paradigm. By carefully making a competition and cooperation trade-off among agents, we can reach an equilibrium state that guarantees not only individual advertiser's utility but also the system performance (i.e., social welfare). Second, to avoid the potential collusion behaviors of bidding low prices underlying the cooperation, we further propose bar agents to set a personalized bidding bar for each agent, and then alleviate the revenue degradation due to the cooperation. Third, to deploy MAAB in the large-scale advertising system with millions of advertisers, we propose a mean-field approach. By grouping advertisers with the same objective as a mean auto-bidding agent, the interactions among the large-scale advertisers are greatly simplified, making it practical to train MAAB efficiently. Extensive experiments on the offline industrial dataset and Alibaba advertising platform demonstrate that our approach outperforms several baseline methods in terms of social welfare and revenue.

Bloom filter is an efficient data structure for filtering negative keys (keys not in a given set) with substantially small space. However, in real-world applications, there widely exist vulnerable negative keys, which will bring high costs if not being properly filtered, especially when positive keys are added/deleted dynamically. To address the problem, we propose SeeSaw Counting Filter (SSCF), which is innovated with encapsulating the vulnerable negative keys into a unified counter array named seesaw counter array, and dynamically modulating (or varying) the applied hash functions to guard the encapsulated keys from being misidentified. Moreover, we propose ada-SSCF to handle the scenarios where the vulnerable negative keys cannot be obtained in advance. We extensively evaluate our SSCF, which shows that SSCF outperforms the cutting-edge filters by 3 × on averages regarding accuracy while ensuring a low operation latency. All source codes are in [2].

Finding special items, like heavy hitters, top-k, and persistent items, has always been a hot issue in data stream processing for web analysis. While data streams nowadays are usually high-dimensional, most prior works focus on special items according to a certain primary dimension and yield little insight into the correlations between dimensions. Therefore, we propose to find special quadratic elements to reveal close correlations. Based on the items mentioned above, we extend our problem to three applications related to heavy hitters, top-k, and persistent items, and design a generic framework DUET to process them. Besides, we analyze the error bound of our algorithm and conduct extensive experiments on four data sets. Our experimental results show that DUET can achieve 3.5 times higher throughput and three orders of magnitude lower average relative error compared with cutting-edge algorithms.

Abstract Crowdsourcing has become increasingly popular in recent years. In order to achieve the optimal task allocation, one of the most important issues is to select more suitable crowdworkers. By leveraging its pervasiveness, social network can be employed as a novel worker recruitment platform. A robust task allocation scheme over the social network could also consider the word-of-mouth (WoM) mode, in which tasks are delivered from workers to workers. In this paper, we discuss an Non-deterministic Polynomial-Hard (NP-Hard) problem, cost-effective and budget-balanced task allocation (CBTA) problem under the WoM mode in social groups. We propose two heuristic algorithms: CB-greedy and CB-local based on greedy strategy and local search technique, respectively. We also prove that the running time of CB-greedy is $O(m^2\log m)$, whereas CB-local utilizing disjoint-set achieves $O(mn\alpha (m, n))$, where $m$ is the number of edges indicating interactions of social groups, $n$ is the number of social groups and $\alpha $ is the inverse Ackerman function. Extensive experiments validate the efficiency and performance of our proposed algorithms.

Abstract The traditional relay deployment problem typically assumes that the locations of users are known and stationary, which is not realistic in practice. The prevalence of mobile devices has made it possible to collect user trajectory and account for user movement while deploying relays. Under this background, a novel problem trajectory-based relay deployment (TBRD) is put forward. This problem considers communication-related metrics and is aimed at maximizing user connection time as users roam through the target area under relay resource constraints, which is more reasonable than the goal of expanding the relay coverage. To figure out the TBRD, we first propose the concept demand nodes (DNs), which are virtual weighted nodes representing the locations where users frequently pass or stay for a long period. Next, we design a Demand Node Generation algorithm that can transform the continuous historical user trajectory into a number of discrete DNs. By generating DNs, we convert the TBRD problem into a demand node coverage (DNC) problem, which is proved to be NP-complete. Followed by that, we introduce an approximation algorithm, named Submodular Iterative Deployment Algorithm, which solves the DNC problem with the approximation factor $1-\frac{1}{\sqrt{e\cdot (1-1/k)}}$, where $e$ is the mathematical constant, and $k$ is the relay number constraint. Finally, five real trajectory datasets are used to evaluate our proposed algorithm, and the simulation results demonstrate that our algorithm can obtain high coverage for users in motion, which can lead to better user experience. In addition, we also analyze the impact of different parameters on the coverage performance, and under this circumstance, we may safely come to the conclusion that our work is at the leading edge to utilize user trajectories for relay deployment in wireless networks.

Practical and accurate localization systems are important to mobile targets that enable promising services such as navigation and augmented reality. With the proliferation of WiFi, existing WiFi-based localization systems have leveraged RSSI, fingerprints, landmarks, time of arrival, or angle of arrival to locate targets, while no related work pays attention to mobile targets themselves. For wheel-driven mobile targets, such as vehicles, bikes, and wheeled robots, we design and implement WheelLoc, a novel WiFi-based localization system leveraging the rotation of wheels. The specially designed WheelLoc hardware is cost-effective and self-powered with the composition of three commercial antennas and a solar cell, which is also easy to be installed on wheels. A hybrid WheelLoc algorithm is further proposed to realize accurate localization in diverse environments, whether the wheel of targets is static or mobile, indoor or outdoor, on flat or bumpy ground. The movements of individual antennas are exploited to emulate linear, cycloid, and circular antenna arrays using a new formulation of Synthetic Aperture Radar (SAR). Extensive experiments are conducted on bikes in the real world. Performance results demonstrate that WheelLoc does not require any user interaction, yet achieves comparable accuracy with the state-of-the-art localization systems using WiFi.

Current distributed file systems are designed to support PB-scale even EB-scale data storage. Metadata service, which manages file attribute information and the global namespace tree, is crucial to system performance. Distributed metadata management, using multiple metadata servers (MDS's) to store metadata, provides effective approaches to alleviate the workload of a single server. However, maintaining good metadata locality and keeping load balancing among MDS's at the same time is a nontrivial problem. To better take advantage of the current distribution of the metadata, in this article, we present the first machine learning based model called DeepHash, which leverages the neural network to learn a locality preserving hashing (LPH) mapping scheme. DeepHash first converts the metadata nodes to feature vectors by the network embedding technology. Due to the absence of training labels, i.e., the hash values of metadata nodes, we design a pair loss function with distinctive characters to train DeepHash, and introduce the sampling strategy to improve the training efficiency. Besides, we propose an efficient algorithm to dynamically balance the workload and adopt the cache model to improve query efficiency. The experiments on the Amazon EC2 platform demonstrate that the DeepHash can preserve the metadata locality meanwhile maintaining a high load balancing, which denotes the effectiveness and efficiency of DeepHash compared with traditional and state-of-the-art schemes.

The emerging paradigm of Network Function Virtualization (NFV) promises to shorten the renewal cycles of network functions and reduce the capital expenses by flexibly deploying virtualized network functions (VNFs) implementation on commodity servers. However, the required resource of each type (CPU, memory, etc.) for the running VNF should be provisioned to guarantee the performance when processing packets. This comes with different deployment cost, especially in a heterogeneous cloud consisting of a large number of network function platforms from various vendors. To optimally operate VNFs, it is necessary for the network operator to dynamically deploy VNFs in the expensive cloud infrastructures. In this article, we initiate the study of minimizing the deployment cost under multi-resource constraints in a heterogeneous cloud. We formulate multi-resource VNF deployment problem (MVDP) as an optimization program and prove its hardness. We propose an offline$(1, d+1)$-bicriteria approximation algorithm and an$(\mathcal {O}(1), \mathcal {O}(n \cdot \log n))$-competitive online algorithm to deploy VNFs in a scalable manner, where$d$is the number of resource types and$n$is the number of required VNFs. Large-scale simulations and DPDK-based OpenNetVM implementation show that our algorithms can reduce the overall cost by 34% and improve the performance in terms of multi-resource allocation.

The ubiquitous needs for extracting insights from data are driving the emergence of service providers to offer predictions given the inputs from customers. During this process, it is important and highly nontrivial for the service providers to generate proofs of honest predictions without leaking the key parameters of their trained models. In addition, the customers are usually unwilling to reveal their sensitive inputs. In this article, we proposed MVP, which enablesMachine learning prediction in aVerifiable andPrivacy preserving fashion. MVP features the properties of polynomial decomposition and prime-order bilinear groups to simultaneously facilitate oblivious evaluation and batch outcome verification while maintaining function privacy and input privacy. We further instantiated MVP with Support Vector Machines (SVMs) and extensively evaluated its performance for the spam detection task on three practical Short Message Service (SMS) datasets. Our analysis and evaluation results reveal that MVP achieves the desired properties while incurring low computation and communication overhead.

Indoor vehicle localization is an underlying technology for realizing Autonomous Valet Parking (AVP), which demands high accuracy and reliability. However, existing localization technologies, such as GPS, WiFi, Bluetooth, suffer from either low availability or high cost, which are not practical in the real world. In order to put AVP into practice, We desperately need an efficient and reliable indoor vehicle localization technology. In this paper, we propose aDeep learning andFM fingerprint map based indoor vehicleLocalization method, namely DeFLoc, which leverages FM signals to achieve accurate and practical indoor localization. In order to reduce the workload of the FM fingerprints collecting process, DeFLoc uses partially uniform sampling to decrease sample data volume and reconstructs the FM fingerprint map from collected incomplete fingerprints precisely using a dedicated deep Convolutional Neural Network (CNN). To alleviate the influence of signal distortions in some FM frequencies, we further design smooth layers in the neural network for improving the accuracy of map reconstruction. Moreover, we devise a continuous vehicle localization algorithm by considering the preferences of vehicle movements to assist us to calibrate localization. We implemented a prototype of DeFLoc and conducted extensive experiments both in simulation and practice. Evaluation results show that our proposed reconstruction model improves accuracy by 40% over conventional matrix completion methods even under the 60% data missing rate. With the precisely reconstructed fingerprint map, DeFLoc achieves over 90% localization accuracy, which indicates DeFLoc can realize accurate and practical indoor vehicle localization.

Online bipartite matching has attracted wide interest since it can successfully model the popular online car-hailing problem and sharing economy. Existing works consider this problem under either adversary setting or i.i.d. setting. The former is too pessimistic to improve the performance in the general case; the latter is too optimistic to deal with the varying distribution of vertices. In this article, we initiate the study of the non-stationary online bipartite matching problem, which allows the distribution of vertices to vary with time and is more practical. We divide the non-stationary online bipartite matching problem into two subproblems, the matching problem and the selecting problem, and solve them individually. Combining Batch algorithms and deep Q-learning networks, we first construct a candidate algorithm set to solve the matching problem. For the selecting problem, we use a classical online learning algorithm, Exp3, as a selector algorithm and derive a theoretical bound. We further propose CDUCB as a selector algorithm by integrating distribution change detection into UCB. Rigorous theoretical analysis demonstrates that the performance of our proposed algorithms is no worse than that of any candidate algorithms in terms of competitive ratio. Finally, extensive experiments show that our proposed algorithms have much higher performance for the non-stationary online bipartite matching problem comparing to the state-of-the-art.

Influence maximization (IM) aims at maximizing the spread of influence by offering discounts to influential users (called seeding). In many applications, due to user’s privacy concern, overwhelming network scale etc., it is hard to target any user in the network as one wishes. Instead, only a small subset of users is initially accessible. Such access limitation would significantly impair the influence spread, since IM often relies on seeding high degree users, which are particularly rare in such a small subset due to the power-law structure of social networks. In this paper, we attempt to solve the limited IM in real-world scenarios by the adaptive approach with seeding and diffusion uncertainty considered. Specifically, we consider fine-grained discounts and assume users accept the discount probabilistically. The diffusion process is depicted by the independent cascade model. To overcome the access limitation, we prove the set-wise friendship paradox (FP) phenomenon that neighbors have higher degree in expectation, and propose a two-stage seeding model with the FP embedded, where neighbors are seeded. On this basis, for comparison we formulate the non-adaptive case and adaptive case, both proven to be NP-hard. In the non-adaptive case, discounts are allocated to users all at once. We show the monotonicity of influence spread w.r.t. discount allocation and design a two-stage coordinate descent framework to decide the discount allocation. In the adaptive case, users are sequentially seeded based on observations of existing seeding and diffusion results. We prove the adaptive submodularity and submodularity of the influence spread function in two stages. Then, a series of adaptive greedy algorithms are proposed with constant approximation ratio. Extensive experiments on real-world datasets show that our adaptive algorithms achieve larger influence spread than non-adaptive and other adaptive algorithms (up to a maximum of 116 percent).

Spatial crowdsensing is a special kind of crowdsourcing which allocates tasks to workers in some special places where workers can sense data for them. Due to the lack of priori information about the quality of workers and the ground truth, selecting the most suitable workers, which can guarantee the quality of the sensing tasks, remains a great challenge. In this paper, we propose a novel framework which can choose the most reliable workers among available workers under constraint budget. We model the quality of workers through two factors, bias and variance, which describe the continuous feature of sensing tasks. Our framework first allocate some calibration tasks to calibrate the bias and then iteratively estimate the workers variance more and more accurately. To choose more reliable workers, we face the exploration and exploitation dilemma. Therefore, we design a novel Multi-Armed Bandit (MAB) algorithm which based on Upper Confidence Bounds (UCB) scheme and combined with a weighted data aggregation scheme to estimate a more accurate ground truth of a sensing task. Futhermore, a dynamic budget allocation algorithm is designed to achieve global optimization. Then, we prove the expected sensing error can be bounded according to the regret bound of the MAB. In simulation experiments, we compare our algorithm with several baselines with real-world data set and it shows the effectiveness in inferring the ground truth with limited budget.

In recent years, social networks have developed rapidly and become an indispensable part of people’s everyday life. Many models try to predict whether some reshare cascades are going to be popular or not, but most of their performances are limited due to the lack of cascades’ information in the early stage. In this paper, we proposeEarly Pattern detection model for Outbreak Cascades(in abbreviation, EPOC) inspired by the survival theory. We use three features to predict cascades’ virality: retweet sequence, follower number sequence, and timestamps of the first tweet which includes both the static and dynamic characteristics of cascades. Utilizing the theory that distributions of both viral and non-viral cascades are Gaussian, we get the boundary between these two kinds of cascades with sufficient proof to testify its rationality. To detect the virality more precisely and earlier, based on hazard functions in the survival theory, we propose two different hazard ceilings to capture the bursting of the cascades. We also provide a series of numerical experiments to analyze impacts of different factors to performance of our model measured by three practical metrics. The results shows that our model could stably outperforms several state-of-art baselines.

The society’s insatiable appetites for personal data are driving the emergence of data markets, allowing data consumers to launch customized queries over the datasets collected by a data broker from data owners. In this paper, we study how the data broker can maximize its cumulative revenue by posting reasonable prices for sequential queries. We thus propose a contextual dynamic pricing mechanism with the reserve price constraint, which features the properties of ellipsoid for efficient online optimization and can support linear and non-linear market value models with uncertainty. In particular, under low uncertainty, the proposed pricing mechanism attains a worst-case cumulative regret logarithmic in the number of queries. We further extend our approach to support other similar application scenarios, including hospitality service and online advertising, and extensively evaluate all three use cases over MovieLens 20M dataset, Airbnb listings in U.S. major cities, and Avazu mobile ad click dataset, respectively. The analysis and evaluation results reveal that: (1) our pricing mechanism incurs low practical regret, while the latency and memory overhead incurred is low enough for online applications; and (2) the existence of reserve price can mitigate the cold-start problem in a posted price mechanism, thereby reducing the cumulative regret.

Predicting emerging hot events in an early stage is essential for various applications, including information dissemination mining, ads recommendation and etc. Existing techniques either require a long-term observation over the event or features that are expensive to extract. However, given limited data at the early stage of an emerging event, the temporal features of hot events and non-hot events are not distinctive enough yet. In this work, we introduce BEEP, a Bayesian perspective Early stage Event Prediction model, that tackles this dilemma. We formulate the hot event prediction problem by two Semi-Naive Bayes Classifiers, where we consider both the temporal features and structural features and perform distribution test for the selected features. Theoretical analysis and extensive empirical evaluations on two real datasets demonstrate the effectiveness of our methods.

One critical issue for wireless power transfer is to avoid human health impairments caused by electromagnetic radiation (EMR) exposure. The existing studies mainly focus on scheduling wireless chargers so that (expected) EMR at any point in the area does not exceed a threshold$R_t$. Nevertheless, they overlook the EMR jitter that leads to exceeding of$R_t$even if the expected EMR is no more than$R_t$. This paper studies the fundamental problem ofRObustlySafEcharging for wireless power transfer (ROSE), that is, scheduling the power of chargers so that the charging utility for all rechargeable devices is maximized while the probability that EMR anywhere does not exceed$R_t$is no less than a given confidence. We first build our empirical probabilistic charging model and EMR model. Then, we present EMR approximation and area discretization techniques to formulate ROSE into a Second-Order Cone Program. After that, we propose the first redundant second-order cone constraints reduction algorithm to reduce the computational cost, and therefore obtain a$(1-\epsilon)$-approximation centralized algorithm. Further, we propose a$(1-\epsilon)$-approximation fully distributed algorithm scalable with network size for ROSE. We conduct both simulation and field experiments, and the results show that our algorithms can outperform comparison algorithms by 480.19 percent.

Sweep coverage plays an important role in many applications such as data gathering, sensing coverage and devices control. In this paper, we deal with the cooperative sweep coverage problem with multiple mobile sensors to periodically cover all positions of interest (PoIs) in the surveillance region. Different from traditional sweep coverage scenarios, the cooperative sweep coverage (CSC) problem allows the deployment of multiple sensors on the same trajectory to further reduce the sweep period or detection delay. We also consider the multi-sink sweep coverage (MSSC) problem where each mobile sensor must periodically transmit its collected data to a base station due to the limited storage capacity and power supply. Correspondingly, we propose two constant-factor approximations, namely${{CoCycle}}$and${{SinkCycle}}$, to minimize the maximum sweep period for these two problems. The approximation ratios of${{CoCycle}}$and${{SinkCycle}}$are proved to be 4 and 6 respectively. As far as we know,${{SinkCycle}}$is the first approximation for the sweep coverage problem with multiple sinks. We also provide two optimal algorithms for the CSC problem in one dimensional case and a useful insight regarding the MSSC problem with only one available sink. Finally, we conduct various numerical experiments to validate the effectiveness and efficiency of our designs.

Many realistic mobile social networks can be characterized by evolving bipartite graphs, in which dynamically added elements are divided into two entities and connected by links between these two entities, such as users and items in recommendation networks, authors and scientific topics in scholarly networks, male and female in dating social networks, etc. However, given the fact that connections between two entities are often weighted, how to mathematically model such weighted evolving bipartite relationships, along with quantitative characterizations, remains unexplored. Motivated by this, we develop a novel evolving bipartite model (EBM), which, based on empirically validated power-law distribution on multiple realistic mobile social networks, discloses that the distribution of total weights of incoming and outgoing edges in networks is determined by the weighting scale and bounded by certain ceilings and floors. Based on these theoretical results, for evolving bipartite networks whose degree follows power-law distribution, their overall weights of vertices can be predicted by EBM. To illustrate, in recommendation networks, the evaluation of items, i.e., total rating scores, can be estimated through the given bounds; in scholarly networks, the total numbers of publications under specific topics can be anticipated within a certain range; in dating social networks, the favorability of male/female can be roughly measured. Finally, we perform extensive experiments on 10 realistic datasets and a synthetic network with varying weights, i.e., rating scales, to further evaluate the performance of EBM, and experimental results demonstrate that given weighting scales, both the upper bound and the lower bound of total weights of vertices in mobile social networks can be properly predicted by the EBM.

In this paper, we study the problem ofRobust offloading schEduling for mobIle edge computiNg (REIN), i.e., in the presence of uncertain offloading failures, how to determine an offloading schedule to minimize the overall latency of all computation-intensive tasks. We mathematically formulate the problem in the form of min-max robust optimization, based on the twice equivalent transformations of the scheduling problem that originally does not consider robustness. REIN is challenging to solve because the min-max robust objective is computationally intractable with existing approaches, and the monotonicity of the objective function is uncertain, even if we transform the objective into the popular max-min form by introducing an appropriate constant upper bound. To solve the above challenges, we first construct a constant upper bound and a monotone modular function to approximate the transformed max-min objective function, and then propose a computationally feasible solution with provable performance bound. Moreover, given the fact of the weak computation ability of users in practical, we construct a tighter constant upper bound and a monotone submodular approximation function, and propose a feasible solution with possibly improved performance bound. Extensive results show that, given a maximum number of offloading failures, our proposed algorithms outperform three benchmark algorithms, and approach the optimum at small time costs.

This paper considers the fundamental problem of deployment of Unmanned AerialVehIcles for aniSotropic monItoringTasks (VISIT), that is, given a set of objects with determined coordinates and directions in 2D area, deploy a fixed number of UAVs by adjusting their coordinates and orientations such that the overall monitoring utility for all objects is maximized. We develop a theoretical framework to address VISIT problem. First, we establish monitoring model whose quality of monitoring (QoM is anisotropic with monitoring angle and varying with various monitoring distance. To the best of our knowledge, we are the first considering the anisotropy of monitoring angle. Then, we propose a framework consisting of area discretization and Monitoring Dominating Set (MDS) extraction to reduce the infinite solution space of VISIT to a limited one with performance bound. Finally, we model the reformulated problem as maximizing a monotone submodular function subject to a matroid constraint, and present a greedy algorithm with$1-1/e-\epsilon$approximation ratio. We conduct both simulations and field experiments to evaluate our framework, and the results show that our algorithm outperforms comparison algorithms by at least 41.3 percent.

As Internet users interacting with their different friends in different social networks, the de-anonymization problem has been raising improving concern. Since the assailants may de-anonymize a social network by matching it with a correlated sanitized network and identifying anonymized user identities, multifarious arts study on the theoretical conditions or practical algorithms for correctly de-anonymizing a social network. Except for the structural information of these social networks, there has also been bounteous works taking advantage of some pre-identified seed nodes for reference in the anonymized network. In this paper, we systematically probe the theoretical conditions and algorithmic approaches for correctly matching two different-sized social networks by leveraging the multi-hop neighborhood relationships. A limited number of seeds are also taken into consideration as auxiliary information. To this end, we introduce the de-anonymization problem with the aid of the collectiveness and the collective adjacency disagreements, which are the collection of disagreements of different multi-hop adjacency matrices. We theoretically demonstrate that minimizing the collective adjacency disagreements can help match two social networks even in a very sparse circumstance, as it significantly enlarges the difference between the mismatched node pairs and the correctly matched pairs. Besides, the seeds is proved to bring positive influence in improving the de-anonymization accuracy. Algorithmically, we relax the domain of the matching function to continuum and adopt the conditional gradient descending method on the collective-form objective, to efficiently minimize the collective adjacency disagreements of two networks. We conduct tremendous experiments on different networks with or without seeds, the results of which return desirable de-anonymization accuracies and reveal the advantages of the collectiveness: the collectiveness manifests rich structural information, thereby most nodes can be correctly matched with their correspondences even in some sparse networks, where merely utilizing the 1-hop adjacency relationships might fail to work.

Mobile crowdsensing has become increasingly popular due to its ability to collect a massive amount of data with the help of many individual smartphone users. A crowdsensing platform can utilize the collected data to extract effective information and provide diverse services. Designing an incentive mechanism to compensate the participants for their resources consumption is critical in attracting more participation. Offline incentive mechanism design has been widely studied in various crowdsensing applications, whereas the online scenario, is much more challenging due to the unavailability of future information when the platform makes user selection decisions. In this paper, we investigate the problem of online crowdsensing by considering a critical property that the values of users’ contributions decrease as time goes by. The time-discounting property is common in inter-temporal choice scenarios but has not been carefully addressed from the perspective of mechanism design. To handle this problem, we propose a new method to select users based on a time-dependent threshold, and present a strategy-proof framework where participants prefer to submit their true types, instead of manipulating the market by misreporting their private information. We consider two cases, one is that the total value is the summation of each participant's contributing value, the other is more general that the total value function is submodular. We call these two mechanisms TDM and TDMS, respectively. We prove that our two mechanisms can achieve computational efficiency, budget feasibility, strategy-proofness, and a constant competitive ratio, in the context of time-discounting values. By comparing our mechanisms with the state-of-the-art methods, we show that our design achieves better performance in terms of the total value.

Anonymization techniques have tranquilized current social network users in terms of privacy leakage, however, it does not radically prevent adversaries from de-anonymizing users, as they may map the users to an un-anonymized network. Till now, researchers share a common thread in such de-anonymization attack: unveiling conditions leading to successful de-anonymization under the chosen network model. However, it has not yet been well understand how the structural property in different network models intrinsically determines de-anonymizability. We address the above issue in this paper by making the two contributions: (i) We discover that the automorphic degree and homomorphic degree of social networks determine their de-anonymizability universally. The automorphic degree characterizes the distinguishability of the users in a network, while the homomorphic degree models the similarities of users between two networks. We conclude that a smaller automorphic degree and a larger homomorphic degree conduce to a higher de-anonymizability. Such model-independent phenomenon refreshes us with a latitudinal study as it generalizes the essential commonness of de-anonymization in different network models. (ii) We derive explicit parametric bounds of the de-anonymizability for three classic network models, showing that such bounds correspond well to our conclusion about morphism property. We then algorithmically and experimentally show that such theoretical results literally make sense to adversaries. Such longitudinal study, including welding theory, algorithm and validation, promises applicability of our results on morphism property in real cases.

Network update enables Software-Defined Networks (SDNs) to optimize the data plane performance. The single update focuses on processing one update event at a time,i.e., updating a set of flows from their initial routes to target routes, but it fails to handle continuously arriving update events in time incurred by high-frequency network changes. On the contrary, the continuous update proposed in “Update Algebra” can handle multiple update events concurrently and respond to the network condition changes at all times. However, “Update Algebra” only guarantees the blackhole-free and loop-free update. The congestion-free property cannot be respected. In this paper, we propose Coeus to achieve the continuous update while maintaining consistency,i.e., ensuring the blackhole-free, loop-free, and congestion-free properties simultaneously. Firstly, we establish the continuous update model based on the update operations in update events. With the update model, we dynamically reconstruct the operation dependency graph (ODG) to capture the relationship between update operations and link utilization variations. Then, we develop a composition algorithm to eliminate redundant operations in update events. To further speed up the update procedure, we present a partition algorithm to split the operation nodes of the ODG into a series of suboperation nodes that can be executed independently. The partition algorithm is proven to be optimal. Finally, extensive evaluations show that Coeus can improve the update speed by at least 179% and reduce redundant operations by at least 52% compared with state-of-the-art approaches when the arrival rate of update events equals three times per second.

To build a secure wireless networking system, it is essential that the cryptographic key is known only to the two (or more) communicating parties. Existing key extraction schemes put the devices into physical proximity and utilize the common inherent randomness between the devices to agree on a secret key, but they often rely on specialized hardware (e.g., the specific wireless NIC model) and have low bit rates. In this article, we seek a key extraction approach that only leverages off-the-shelf mobile devices, while achieving significantly higher key generation efficiency. The core idea of our approach is to exploit the fast varying inaudible acoustic channel as the common random source for key generation and wireless parallel communication for exchanging reconciliation information to improve the key generation rate. We have carefully studied and validated the feasibility of our approach through both theoretical analysis and a variety of measurements. We implement our approach on different mobile devices and conduct extensive experiments in different real scenarios. The experiment results show that our approach achieves high efficiency and satisfactory robustness. Compared with state-of-the-art methods, our approach improves the key generation rate by 38.46% and reduces the bit mismatch ratio by 42.34%.

Datacenter networks enable multiple paths between hosts to provide large bisection bandwidth. It requires load balancers to cope with network uncertainties such as traffic dynamics and topology asymmetry. Existing edge-based load balancing schemes are usually faced with the problem of limited network visibility. This article proposesMeet, a rack-level pooling based load-balancer deployed at the edge that can handle the aformentioned uncertainties.Meetutilizes both passive information as well as active probing to comprehensively sense the network conditions with relatively low cost.Meetdynamically reroutes flows effectively based on the visibility of the network condition.Meethas been tested with extensive flow-level simulations against state-of-the-art load balancers. It outperforms Hermes by up to 10% in the experiments, and outperforms others solutions such as DRILL by up to 50%.Meetrequires no modifications to the switches and is feasible to deploy at the edge.

Deep learning (DL) is becoming increasingly popular in many domains, including computer vision, speech recognition, self-driving automobiles, etc. GPU can train DL models efficiently but is expensive, which motivates users to share GPU resource to reduce money costs in practice. To ensure efficient sharing among multiple users, it is necessary to develop efficient GPU resource management and scheduling solutions. However, existing ones have several shortcomings. First, they require the users to specify the job resource requirement which is usually quite inaccurate and leads to cluster resource underutilization. Second, when scheduling DL jobs, they rarely take the cluster network characteristics into consideration, resulting in low job execution performance. To overcome the above issues, we propose Liquid, an efficient GPU resource management platform for DL jobs with intelligent resource requirement estimation and scheduling. First, we propose a regression model based method for job resource requirement estimation to avoid users over-allocating computing resources. Second, we propose intelligent cluster network-efficient scheduling methods in both immediate and batch modes based on the above resource requirement estimation techniques. Third, we further propose three system-level optimizations, including pre-scheduling data transmission, fine-grained GPU sharing, and event-driven communication. Experimental results show that our Liquid can accelerate the job execution speed by 18% on average and shorten the average job completion time (JCT) by 21% compared with cutting-edge solutions. Moreover, the proposed optimization methods are effective in various scenarios.

The average/tail Flow Completion Times (FCTs) are critical to many datacenter applications. Congestion control plays a central role in optimizing FCT. Inappropriate congestion control can exacerbate buffer occupancy, thus hurting the flow performance. Our observations are that current approaches are too aggressive in injecting packets into underlying networks. Instead of handling buffer explosion afterward, we reduce buffer occupancy in the first place. We propose PayDebt, a novel and readily-deployable proactive congestion control protocol. At its heart, adebtmechanism provides bandwidth coordination between the already-buffered and the forthcoming packets. We evaluate PayDebt both in a testbed and large-scale simulations. The buffer occupancy can be decreased by up to 8.0×-35.9× compared to DCQCN and Homa.

To minimize a job's completion time, we need to minimize the completion time of its final stage's last task. Scheduling of machine slots and networks largely dominates the variable part of each task's duration. Finding an optimal schedule is NP-hard even for offline and simplified scenarios. Previous work does lead to improved performance with various strategies. State-of-the-art task placement and network scheduling efforts are largely disjunctive. Without joint optimization, they are sub-optimal and myopic in many scenarios. Task placement usually treats the network as a black box. Thus, we use prioritized bandwidth allocation among tasks making the network bothpredictableandefficientto achieve joint scheduling. With this feature, joint scheduling can be transformed into a specialbin-packing problem. Over this minimal yet power-enough abstraction, we propose PushBox to schedule data-parallel jobs in multi-tenant clusters. When designing the joint scheduling algorithm, we not only embrace the wisdom of prior art but also respect administrators’ fairness intent, which is so far largely ignored. We implement PushBox on Hadoop 3. PushBox performs persistently well on both a small testbed and a trace-driven simulator.

The advance of tour recommendation allows people to get well-fit route plans, which contain a sequence of Points of Interest (POIs) based on tourists’ constraints and preferences. Large-scale POIs, called Super-POIs in this article, often contain multiple scenic spots and entrances. Tourists have to specify a suitable tour route inside Super-POI to obtain good tour experience. However, most of existing tour recommendation algorithms ignore the detailed information inside Super-POIs. By taking super-POIs into account, we proposeEmbedded Tour(eTOUR), a two-layer framework considering route design of POIs (Outer Model) and scenic routes inside the Super-POIs (Inner Model) respectively. To combine two models, an Embedded GRASP-VNS Algorithm is introduced based on an embedding strategy. For Outer Model, we apply Greedy Randomized Adaptive Search Procedure (GRASP) for route construction and Variable Neighborhood Search (VNS) for local improvement. Super-POI is treated as a “meta node” in outer route construction. For Inner Model, the optimal route inside Super-POI obtained by DFS-based Tree Search with Pruning is revised dynamically to adapt to the outer route. Furthermore, we discuss a special case in the Super-POI where a key graph is defined and treated as the “must go” route. We modify the solution of Chinese Postman Problem in this case to reduce the time complexity. Finally, experiments based on two real datasets demonstrate the effectiveness of our proposal.

A large-scale and high-quality image collection is a fundamental demand in the 3D reconstruction scenario. Crowdsourcing can help us to collect lots of diversified images. However, it is difficult to attract people to accomplish tasks due to their self-interest. Besides, the quality of collected images is various. Low-quality images may degrade the performance of 3D reconstruction. To avoid low-quality images and motivate participants to provide high-quality images, we take image quality into account when allocating rewards. In this article, we propose a pricing mechanism, called ImgPricing, to determine the rewards of participants in 3D reconstruction. We model the process of image collection as a cooperative game, and regard image quality and the arrival sequence of images as critical factors in the reward allocation. ImgPricing differs from traditional pricing schemes, e.g., Shapley value and Banzhaf power index-based methods, in that it introduces the images’ arrival sequence to be an indispensable element. We lastly implement ImgPricing on the Android platform and extensively evaluate its performance. Our evaluation results demonstrate that ImgPricing outperforms other existing schemes in terms of computational efficiency, fairness, and robustness. In brief, our image quality-based pricing mechanism for crowdsourced 3D reconstruction is feasible and effective.

Network measurement provides operators an efficient tool for many network management tasks such as performance diagnosis, traffic engineering and intrusion prevention. However, with the rapid and continuous growth of traffic speed, it needs more computing and memory resources to monitor traffic in per-flow or per-packet granularity. Sample-based measurement systems (e.g., NetFlow, sFlow) have been developed to perform coarse-grained measurement, but they may miss part of records, especially for mice flows, which are important for some network management tasks (e.g., anomaly detection, performance diagnosis). To address these issues, data streaming algorithms such as hash tables and sketches have been introduced to balance the trade-off among accuracy, speed, and memory usage. In this article, we present a systematic survey of various data structures, algorithms and systems which have been proposed in recent years to perform fine-grained measurement for high-speed networks. We organize these methods and systems from a software-defined perspective. In particular, we abstract fine-grained network measurement into three-layer architecture. We introduce the responsibility of each layer and categorize existing state-of-the-art works into this architecture. Finally, we conclude the article and discuss the future directions of fine-grained network measurement.

In urban fields, Mobile Wireless Sensor Networks (MWSNs) become ubiquitous. Accurate GPS positioning for sensors is a fundamental problem for MWSNs. To solve this problem, this paper proposes a Crowdsourcing-Aided Positioning scheme, which takes an ideal situation and a more realistic situation into account. In the ideal situation, all participants are considered accurate. Then, two optimization objectives are addressed for the efficient Crowdsourcing-Aided Positioning task. Their utility functions are proven to be submodular and a greedy algorithm is given to solve them. In the more realistic situation, randomly selected participants cannot guarantee the accuracy of the data. We propose a data-accuracy-calibration-based participant selection framework to solve this dilemma. Through data accuracy calibration, participants gain their data accuracy and reliability with the help of wireless sensor networks. First, we design three kinds of data accuracy calibration methods based on probabilistic models. Then, we propose a Truthful-Data-Driven Participant Selection problem, which tends to raise the data accuracy and reliability. The optimization problem is proved to be NP-hard and its optimization function has submodular property. We give a greedy algorithm with$1-\frac {1}{e}$approximation ratio to solve this problem. Simulation experiments are conducted to validate the algorithmic effectiveness at last.

Federated learning allows mobile clients to jointly train a global model without sending their private data to a central server. Extensive works have studied the performance guarantee of the global model, however, it is still unclear how each individual client influences the collaborative training process. In this work, we defined a new notion, called {\em Fed-Influence}, to quantify this influence over the model parameters, and proposed an effective and efficient algorithm to estimate this metric. In particular, our design satisfies several desirable properties: (1) it requires neither retraining nor retracing, adding only linear computational overhead to clients and the server; (2) it strictly maintains the tenets of federated learning, without revealing any client's local private data; and (3) it works well on both convex and non-convex loss functions, and does not require the final model to be optimal. Empirical results on a synthetic dataset and the FEMNIST dataset demonstrate that our estimation method can approximate Fed-Influence with small bias. Further, we show an application of Fed-Influence in model debugging.

User behavior sequences contain rich information about user interests and are exploited to predict user's future clicking in sequential recommendation. Existing approaches, especially recently proposed deep learning models, often embed a sequence of clicked items into a single vector, i.e., a point in vector space, which suffer from limited expressiveness for complex distributions of user interests with multi-modality and heterogeneous concentration. In this paper, we propose a new representation model, named as Seq2Bubbles, for sequential user behaviors via embedding an input sequence into a set of bubbles each of which is represented by a center vector and a radius vector in embedding space. The bubble embedding can effectively identify and accommodate multi-modal user interests and diverse concentration levels. Furthermore, we design an efficient scheme to compute distance between a target item and the bubble embedding of a user sequence to achieve next-item recommendation. We also develop a self-supervised contrastive loss based on our bubble embeddings as an effective regularization approach. Extensive experiments on four benchmark datasets demonstrate that our bubble embedding can consistently outperform state-of-the-art sequential recommendation models.

With the increase of diversity in application needs and networks, existing congestion control algorithms (CCAs) do not accommodate this complicated reality. Previous classic CCAs are designed for a specific domain with fixed rules, failing to adapt to such diversities. Recently surged learning-based CCAs have great potential in adaptability and flexibility but are not practical due to unsatisfying performance on convergence, fairness, overhead and safety assurance. In this paper, we propose Libra, a unified congestion control framework, which empowers flexibility, adaptability, and practicality, by combining the wisdom of classic and reinforcement learning (RL)-based CCAs. Extensive evaluation of Libra's Linux kernel implementations on both live Internet and emulated networks shows performance improvement under dynamic networks (e.g., 1.2x throughput than Orca on average). At the same time, Libra can flexibly satisfy different application needs, reduce the running overhead by at most 0.92x and perform good fairness and convergence properties, well-fitting our theoretical analysis.

Incast occurs frequently in datacenter networks where a large number of senders send data to a single receiver simultaneously, which makes the last hop the network bottleneck. Incast can hurt flows' performance. However, congestion control protocols are not effective at handling incast. One key insight is that it is too late to handle incast packets after they have already piled up at the last hop. Instead, we should avoid incast as early as possible. Inspired by flood control in Hydrologic Engineering, we propose Floodgate, a novel switch-based per-hop flow control to handle incast. Floodgate is compatible with existing congestion control protocols. We integrate it with practical congestion control approaches such as DCQCN, TIMELY, and HPCC. We evaluate Floodgate both in our implementations and large-scale simulations. Compared with state of the art, Floodgate reduces the buffer occupancy by a factor of 6.6x, as well as the queuing delay. Therefore, the average FCT and tail latency are greatly reduced.

Internet of Vehicles (IoV) has become an indispensable technology to bridge vehicles, persons and infrastructures, and is promising to make our cities smarter and more connected. It enables vehicles to exchange vehicular data (e.g., GPS, sensors, and brakes) with different entities nearby. However, sharing these vehicular data over the air raises concerns about identity privacy leakage. Besides, the centralized architecture adopted in existing IoV systems is fragile to single point of failure and malicious attacks. With the emergence of blockchain technology, it has the chance to solve these problems due to its features of tamper-proof, traceability and decentralization. In this paper, we propose a privacy-preserving vehicular data sharing framework based on blockchain. In particular, we design an anonymous and auditable data sharing scheme using Zero-Knowledge Proof (ZKP) technol-ogy so as to protect the identity privacy of vehicles while preserving the vehicular data auditability for Trusted Authorities (TAs). In response to high mobility of vehicles, we design an efficient multi-sharding protocol to decrease blockchain communication costs without compromising the blockchain security. We implement a prototype of our framework and conduct extensive experiments and simulations on it. Evaluation and analysis results indicate that our framework can not only strengthen system security and data privacy, but also increase the data authenticity verification efficiency by 5x comparing to existing privacy-preserving schemes.

Bloom filter is a compact memory-efficient probabilistic data structure supporting membership testing, i.e., to check whether an element is in a given set. However, as Bloom filter maps each element with uniformly random hash functions, few flexibilities are provided even if the information of negative keys (elements are not in the set) are available. The problem gets worse when the misidentification of negative keys brings different costs. To address the above problems, we propose a new Hash Adaptive Bloom Filter (HABF) that supports the customization of hash functions for keys. The key idea of HABF is to customize the hash functions for positive keys (elements are in the set) to avoid negative keys with high cost, and pack customized hash functions into a lightweight data structure named HashExpressor. Then, given an element at query time, HABF follows a two-round pattern to check whether the element is in the set. Further, we theoretically analyze the performance of HABF and bound the expected false positive rate. We conduct extensive experiments on representative datasets, and the results show that HABF outperforms the standard Bloom filter and its cutting-edge variants on the whole in terms of accuracy, construction time, query time, and memory space consumption (Note that source codes are available in [1]).

Multipath TCP (MPTCP) is a burgeoning transport protocol which enables the server to split the traffic across multiple network interfaces. Classic MPTCPs have good friendliness and practicality such as relatively low overhead, but are hard to achieve consistent high-throughput and adaptability, especially for the ability of flexibly balancing congestion among different paths. In contrast, learning-based MPTCPs can essentially achieve consistent high-throughput and adaptability, but have poor friendliness and practicality. In this paper, we proposed MPLibra, a combined multipath congestion control framework that can complement the advantages of classic MPTCPs and learning-based MPTCPs. Extensive simulations on NS3 show that MPLibra can achieve good performance and outperform state-of-the-art MPTCPs under different network conditions. MPLibra improves the throughput by 40.5% and reduces the file download time by 47.7% compared with LIA, achieves good friendliness and balances congestion timely.

Edge computing provides task offloading services to extend the computational capacity of mobile users and reduce task latency. The deadline-awareness offloading algorithm plays a key role in guaranteeing the quality of service (QoS) requirement. Prior studies mainly focus on tasks with strict deadlines. However, some tasks may not always have to be finished before hard deadlines, e.g., multimedia tasks. Tasks with soft deadlines can miss their primary deadlines, but not by too much. This has not been properly considered by existing offloading approaches. In this paper, we propose CONFECT to offload tasks with mixed deadlines. We formulate the problem and prove its hardness. Then, we propose two online algorithms with proven competitive ratios to solve the problem collectively, including an algorithm that assigns tasks to edge servers and an algorithm that adjusts the task execution order on each server. Extensive experiments show that CONFECT outperforms five baseline algorithms.

The development of global positioning system stimulates the popularity of location-based social network (LBSN) services. With a large volume of data containing locations, texts, check-in information, and social relationships, spatial keyword queries in LBSNs have become increasingly complex. In this paper, we identify and solve the Social-based Time-aware Spatial Keyword Query (STSKQ) that returns the top-k objects by considering geo-spatial score, keywords similarity, visiting time score, and social relationship effect. To tackle STSKQ, we propose a two-layer hybrid index structure called Network Embedding Time-aware R-tree (NETR-Tree). In the user layer, we exploit the network embedding strategy to measure the relationship effect in users' relationship network. In the location layer, we build a Time-aware R-tree (TR-tree) considered spatial objects' spatiotemporal check-in information, and present a corresponding query processing algorithm. Finally, extensive experiments on two different real-life LBSNs demonstrate the effectiveness and efficiency of our methods, compared with existing state-of-the-art methods.

With the ubiquity of GPS-enabled smart devices equipped with high-fidelity sensors and increased availability of the wireless network, spatial crowdsourcing has been recently proposed as a general framework to employ smart device carriers as workers to provide services and perform location-sensitive tasks. In this paper, we focus on the task assignment in spatial crowdsourcing, which aims to find the optimal strategy to assign each task to a proper worker such that the total number of completed tasks is maximized and the traveling time cost is minimized, while the workers can return to their initial locations before deadlines after performing the assigned tasks. Finding the optimal global assignment turns out to be intractable since it does not simply imply optimality for an individual worker, as a typical nearest-neighbor heuristic does not render a satisfactory result in general. In spatial crowdsourcing, we model the task assignment problem as a multiple objective joint optimization problem, which focuses on maximizing accomplished task rate and minimizing travel time cost rate simultaneously, and solves it with a mixed priority queue scheduling algorithm. We also introduce a spectral clustering algorithm in spatial crowdsourcing for the first time to divide the task network into different subdomains, considering the task clustering phenomena in real scenarios. Experiments on synthetic and real-world networks demonstrate the efficiency and effectiveness of our method in the task assignment of spatial crowdsourcing and provide insights into its application in practice.

Travel route recommendation aims to recommend a sequence of point of interests (POIs) for visitors based on their personal interests. Previous studies utilize user interest features and POI spatial information to provide travel route recommendation service. However, most of them fail to consider the implicit information in user traveling patterns, which is the key to improve recommendation quality. Additionally, few deep learning based travel route recommendation systems provide comprehensive trip planning functionalities, which is critical to improve the user experience. To alleviate these two problems, we propose a multi-functional attention-based neural network for route recommendation (named MatTrip). We first introduce an encoder-decoder structure with a novel dual bi-directional LSTM encoder as the sequence generation model to learn other users' traveling records and generates a semantic travel route based on user preference and geographical features of start/end POI. Next, multiple user-specific functionalities are supported in MatTrip by grid beam search. The functionalities include weather dependency, POI opening hours, restricted sequence length, mandatory POIs, and dynamic route revision. In addition, MatTrip adopts an online learning approach to learn from user deviation behaviors to improve recommendation performance. Experiments on two real-world datasets show that our model achieves a 20.98% improvement in performance, compared with state-of-arts.

Continuous domestic respiration monitoring provides vital information for diagnosing assorted diseases. In this paper, we introduce RESPTRACKER, the first continuous, multiple-person respiration tracking system in domestic settings using acoustic-based COTS devices. RESPTRACKER uses a two-stage algorithm to separate and recombine respiration signals from multiple paths in a short period so that it can track the respiration rate of multiple moving subjects. Our experimental results show that our two-stage algorithm can distinguish the respiration of at least four subjects at a distance of three meters.

This paper studies social network de-anonymization problem, which aims to identify users of an anonymized network by matching its user set with that of another auxiliary sanitized network. Prior arts primarily assume that both networks share exactly the same set of users, as opposed to many real situations of partially shared users in between. Different from the full matching case that only needs to take care of increasing the number of correctly matched pairs, the case of partial overlapping imposes additional demand on avoiding the wrong matches of those who do not have accounts across networks.To this end, we establish a new cost function, which we call the structural F-score to incorporate both the structural commonness and difference across networks. Intrinsically, the structural F-score computes the ratio of link agreements and disagreements, thus serving as the harmonic mean of precision and recall for any given matching function. Theoretically, we show that for networks parameterized by node overlap t2and link overlap s2, as long as the mean degree of networks grows as Ω(t-2s-3log n), maximizing the structural F-score provably ensures the perfect matching, where the nodal precision and recall are both maximized to 1. Algorithmically, for small-scale networks, we propose a two-step heuristic of F-score based de-anonymization, which firstly finds the optimal full matching between networks and then removes those pairs hindering structural F-score maximization. Due to the universal adaptability of the structural F-score, we further extend the algorithm to large-scale networks via a progressive matching process. Empirical results also validate the effectiveness of our methods in terms of improving the nodal F-score.

State-of-the-art inter-datacenter WANs rely on centralized traffic engineering (TE) to improve the network performance, where TE computation is a periodical procedure and timely performs routing configurations (i.e., enforces TE polices via add, remove and modify forwarding rules) in response to the changing network conditions. The TE computation determines the routing configurations corresponding to the current network conditions and the network update operations change the routing configurations from last TE to current TE solution. Existing works take centralized TE computation and network update as two individual optimization procedures, which inevitably leads to suboptimal solution in the long run. In this paper we initiate the study of online joint optimization on TE computation and network update with the objective of minimizing the sum of TE cost and network update cost. We formulate this problem as an optimization program and propose a set of provable online algorithms with rigorous competitive and regret analysis. Trace-driven simulations on two empirical topologies demonstrate that our algorithms can significantly decrease the total cost.

In online advertising, it is important to jointly consider multiple objectives, e.g. platform revenue and display quality. The existing work only considered multi-objective optimization in each single auction stage, which can be suboptimal in the setting with multiple stages in real online advertising. In this paper, we propose a dynamic auction mechanism which can make a tradeoff between revenue and quality across a series of multiple auction stages, to further improve the performance of the Pareto frontiers from the optimal static auction mechanism in each single stage. We prove that the proposed dynamic auction mechanism satisfies dynamic incentive properties. As the exact valuation distributions are usually unavailable in practice, we further propose a practical implementation of our multi-objective dynamic auction mechanism via data-driven methods and with guarantee of approximate dynamic incentive compatibility. Finally, we confirm our theoretical results and evaluate our practical implementation via empirical study on both synthetic data and real industrial data, and observe a significant improvement on both revenue and quality than the static auction mechanism baseline.

Campus networks are widely deployed for organizations like universities and large companies. Applications and network-based services require campus networks to guarantee short queue and provide low latency and large bandwidth. However, the widely adopted packet-loss-based congestion control mechanism in client hosts builds up long queues in the switch buffer, which is prone to packet loss in burst scenarios, resulting in great network delay. Therefore, a scheme for efficiently controlling queue length of shallow buffer switches in campus networks is urgently needed. Explicit Congestion Notification(ECN) as an explicit feedback mechanism is widely adopted in data center networks to build lossless networks. In this paper, we propose Clean, an efficient queue length control scheme based on transparent ECN-proxy for campus networks. Clean is able to exert fine-grained control over arbitrary client TCP stacks by enforcing per-flow congestion control in the access point(AP). It allows the campus network switches to maintain a low queue length, resulting in high throughput, low latency and zero packet loss. Evaluation results demonstrate that Clean reduces the maximum queue length of the switch by 86% and reduces the 99th percentile latency by 85%. Clean also achieves zero packet loss in burst scenarios.

In e-commerce advertising, it is crucial to jointly consider various performance metrics, e.g., user experience, advertiser utility, and platform revenue. Traditional auction mechanisms, such as GSP and VCG auctions, can be suboptimal due to their fixed allocation rules to optimize a single performance metric (e.g., revenue or social welfare). Recently, data-driven auctions, learned directly from auction outcomes to optimize multiple performance metrics, have attracted increasing research interests. However, the procedure of auction mechanisms involves various discrete calculation operations, making it challenging to be compatible with continuous optimization pipelines in machine learning. In this paper, we design Deep Neural Auctions (DNAs) to enable end-to-end auction learning by proposing a differentiable model to relax the discrete sorting operation, a key component in auctions. We optimize the performance metrics by developing deep models to efficiently extract contexts from auctions, providing rich features for auction design. We further integrate the game theoretical conditions within the model design, to guarantee the stability of the auctions. DNAs have been successfully deployed in the e-commerce advertising system at Taobao. Experimental evaluation results on both large-scale data set as well as online A/B test demonstrated that DNAs significantly outperformed other mechanisms widely adopted in industry.

In-air gesture control extends a touch screen and enables contactless interaction, thus has become a popular research direction in the past few years. Prior work has implemented this functionality based on cameras, acoustic signals, and Wi-Fi via existing hardware on commercial devices. However, these methods have low user acceptance. Solutions based on cameras and acoustic signals raise privacy concerns, while WiFi-based solutions are vulnerable to background noise. As a result, these methods are not commercialized and recent flagship smartphones have implemented in-air gesture recognition by adding extra hardware on-board, such as mmWave radar and depth camera. The question is, can we support in-air gesture control on legacy devices without any hardware modifications?

Meta-learning based recommendation systems alleviate the cold-start problem through a bi-level meta-optimization process. Recommendation borrows prior experience from pre-trained static system-level parameters and fine-tunes the model in user-level for new users. However, it is more natural for the system to sample users in a dynamic online sequence in most real-world recommendation systems, which brings further challenges for existing meta-learning based recommendation: system-level updates begins before user-level recommendation models have converged on the whole time series; stable and randomness-resistant bi-level gradient descent approaches are missing in the current meta-learning framework; evaluation on learning abilities across different users are lacked for exploring the diversities of different users.

Area coverage is one of the most common and important tasks for flying ad hoc networks (FANETs). The increasingly large scale of FANETs brings challenges in communication and coverage. Clustering is an effective technique for networking and management for large-scale ad hoc networks. Meanwhile, some applications, i.e., face recognition, need to perform intensive computation after unmanned aerial vehicles (UAVs) perform area coverage. Due to long response delay in transferring data to the cloud, it becomes a trend to use mobile-edge computing (MEC) for processing data in FANETs, which selects the node of rich computing resources, i.e., cluster head (CH), as MEC server, thus the delay performance of the edge node to the server is particularly critical. However, there is a conflict between area coverage efficiency and delay performance. Area coverage expects UAVs to spread as widely as possible, which may lead to a longer delay. In this article, we consider maximizing coverage efficiency under delay constraints. We define the coverage efficiency and propose an iterative coverage-efficient clustering algorithm (CECA) by applying penalty and block coordinate descent methods. Specifically, the CHs, positions and transmit powers are alternately optimized in each iteration. In addition, CECA can adjust delay constraints according to task requirements. Extensive simulation results show that our proposed approach is superior to other approaches in terms of coverage efficiency and delay.

Real-time information processing is critical to the success of diverse applications from many areas. Age of Information (AoI), as a new metric, has received considerable attention to evaluate the performance of real-time information processing systems. In recent years, edge computing is becoming an efficient paradigm to reduce the AoI and to provide the real-time services. Considering the substantial deployment cost and the resulting resource limitation in edge computing, a proper pricing mechanism is highly necessary to fully utilize edge resources and then minimize the overall AoI of the whole system. However, there are two challenges to design this mechanism: 1) the priorities (or values) of the real-time computing tasks, critical to the efficient resource allocation, are usually private information of users and may be manipulated by selfish users for their own interests; 2) due to the time-varying property of AoI, the values of the tasks discount with time, making the traditional pricing mechanisms infeasible. In this paper, we extend the classical Myerson Theorem to the online setting with time discounting tasks values, and accordingly propose an online auction mechanism, called PreDisc, including an allocation rule and a payment rule. We leverage dynamic programming to greedily allocate resources in each time slot, and charge the winning user with a new critical price, extended from the classical Myerson payment rule. A preemption factor is further employed to make a trade-off between the newly arrived tasks and ongoing tasks. We prove that PreDisc guarantees the economic property of strategy-proofness and achieves a constant competitive ratio. We conduct extensive simulations and the results demonstrate that PreDisc outperforms the traditional mechanisms, in terms of both weighted AoI and revenue of edge service providers. Compared with the optimal solution in offline VCG mechanism, PreDisc has much lower computation complexity with only a slight performance loss.

Although the logically-centralized perspective is offered in Software-Defined Networking (SDN), the data plane is still distributed. Update commands sent by the centralized controller are executed asynchronously and independently in each switch. The timed SDN enables synchronous and coordinated update operations as each update command can be triggered by a pre-defined timestamp. Prior work on timed update mainly focuses on producing a congestion-free update sequence, whereas finding a congestion-free timed update sequence may prolong the update time. Even worse, such an update order may not exist. In this paper, we propose Chronus+, a novel timed update system that utilizes the switch buffer to shorten the update time while minimizing the switch buffer during updates. First, we formulate the Minimum Switch Buffer Problem (MSBP) as an integer linear programming and show its hardness. Then, we propose a set of efficient algorithms to solve the problem in polynomial time. Extensive evaluations in Mininet and large-scale simulations show that Chronus+can reduce the update time by at least 17% and the switch buffer size by at least 27% compared with state-of-the-art approaches.

Online social networks gain increasing popularity in recent years. In online social networks, trust prediction is significant for recommendations of high reputation users as well as in many other applications. In the literature, trust prediction problem can be solved by several strategies, such as matrix factorization, trust propagation, and -NN search. However, most of the existing works have not considered the possible complementarity among these mainstream strategies to optimize their effectiveness and efficiency. In this article, we propose a novel trust prediction approach named iSim : an integrated time-aware similarity-based collaborative filtering approach leveraging on user similarity, which integrates three kinds of factors to measure user similarity, including vector space similarity, time-aware matrix factorization, and propagated trust. This article is the first work in the literature employing time-aware matrix factorization and propagated trust in the study of similarity. Additionally, we use several methods like adding inverted index to reduce the time complexity of iSim , and provide its theoretical time bound. Moreover, we also provide the detailed overview and theoretical analysis of the existing works. Finally, the extensive experiments with real-world datasets show that iSim achieves great improvement for both efficiency and effectiveness over the state-of-the-art approaches.

Influence Maximization (IM) problem is to select influential users to maximize the influence spread, which plays an important role in many real-world applications such as product recommendation, epidemic control, and network monitoring. Nowadays multiple kinds of information can propagate in online social networks simultaneously, but current literature seldom discuss about this phenomenon. Accordingly, in this article, we propose Multiple Influence Maximization (MIM) problem where multiple information can propagate in a single network with different propagation probabilities. The goal of MIM problems is to maximize the overall accumulative influence spreads of different information with the limit of seed budget . To solve MIM problems, we first propose a greedy framework to solve MIM problems which maintains an -approximate ratio. We further propose parallel algorithms based on semaphores, an inter-thread communication mechanism, which significantly improves our algorithms efficiency. Then we conduct experiments for our framework using complex social network datasets with 12k, 154k, 317k, and 1.1m nodes, and the experimental results show that our greedy framework outperforms other heuristic algorithms greatly for large influence spread and parallelization of algorithms reduces running time observably with acceptable memory overhead.

With the increase of mobile devices, Mobile Crowdsensing (MCS) has become an efficient way to ubiquitously sense and collect environment data. Comparing to traditional sensor networks, MCS has a vital advantage that workers play an active role in collecting and sensing data. However, due to the openness of MCS, workers and sensors are of different qualities. Low quality sensors and workers may yield noisy data or even inaccurate data. Which gives the importance of inferring the quality of workers and sensors and seeking a valid task assignment with enough total qualities for MCS. To solve the problem, we adopt truth inference methods to iteratively infer the truth and qualities. Based on the quality inference, this paper proposes a task assignment problem called quality-bounded task assignment with redundancy constraint (QTAR). Different from traditional task assignment problem, redundancy constraint is added to satisfy the preliminaries of truth inference, which requires that each task should be assigned a certain or more amount of workers. We prove that QTAR is NP-complete and propose a (2+ε) - approximation algorithm for QTAR, called QTA. Finally, experiments are conducted on both synthesis data and real dataset. The results of the experiments prove the efficiency and effectiveness of our algorithms.

Predicting the popularity of a single tweet is useful for both users and enterprises. However, adopting existing topic or event prediction models cannot obtain satisfactory results. The reason is that one topic or event that consists of multiple tweets, has more features and characteristics than a single tweet. In this article, we propose two variations of Heterogeneous Bass models (HBass), originally developed in the field of marketing science, namely Spatial-Temporal Heterogeneous Bass Model (ST-HBass) and Feature-Driven Heterogeneous Bass Model (FD-HBass), to predict the popularity of a single tweet at the early stage and the stable stage. We further design an Interaction Enhancement to improve the performance, which considers the competition and cooperation from different tweets with the common topic. In addition, it is often difficult to depict popularity quantitatively. We design an experiment to get the weight of favorite, retweet and reply, and apply the linear regression to calculate the popularity. Furthermore, we design a clustering method to bound the popular threshold. Once the weight and popular threshold are determined, the status whether a tweet will be popular or not can be justified. Our model is validated by conducting experiments on real-world Twitter data, and the results show the efficiency and accuracy of our model, with less absolute percent error and the best Precision and F-score. In all, we introduce Bass model into social network single-tweet prediction to show it can achieve excellent performance.

With the commoditization of personal privacy, pricing private data has become an intriguing problem. In this paper, we study noisy aggregate statistics trading from the perspective of a data broker in data markets. We thus propose ERATO, which enables aggrEgate statistics pRicing over privATe cOrrelated data. On one hand, ERATO guarantees arbitrage freeness against cunning data consumers. On the other hand, ERATO compensates data owners for their privacy losses using both bottom-up and top-down designs. We further apply ERATO to three practical aggregate statistics, namely weighted sum, probability distribution fitting, and degree distribution, and extensively evaluate their performances on MovieLens dataset, 2009 RECS dataset, and two SNAP large social network datasets, respectively. Our analysis and evaluation results reveal that ERATO well balances utility and privacy, achieves arbitrage freeness, and compensates data owners more fairly than differential privacy based approaches.

The cascading of sensitive information such as private contents and rumors is a severe issue in online social networks. One approach for limiting the cascading of sensitive information is constraining the diffusion among social network users. However, the diffusion constraining measures limit the diffusion of non-sensitive information diffusion as well, resulting in the bad user experiences. To tackle this issue, in this paper, we study the problem of how to minimize the sensitive information diffusion while preserve the diffusion of non-sensitive information, and formulate it as a constrained minimization problem where we characterize the intention of preserving non-sensitive information diffusion as the constraint. We study the problem of interest over the fully-known network with known diffusion abilities of all users and the semi-known network where diffusion abilities of partial users remain unknown in advance. By modeling the sensitive information diffusion size as the reward of a bandit, we utilize the bandit framework to jointly design the solutions with polynomial complexity in the both scenarios. Moreover, the unknown diffusion abilities over the semi-known network induce it difficult to quantify the information diffusion size in algorithm design. For this issue, we propose to learn the unknown diffusion abilities from the diffusion process in real time and then adaptively conduct the diffusion constraining measures based on the learned diffusion abilities, relying on the bandit framework. Extensive experiments on real and synthetic datasets demonstrate that our solutions can effectively constrain the sensitive information diffusion, and enjoy a 40 percent less diffusion loss of non-sensitive information comparing with four baseline algorithms.

In mobile crowdsensing, finding the best match between tasks and users is crucial to ensure both the quality and effectiveness of a crowdsensing system. Existing works usually assume a centralized task assignment by the crowdsensing platform, without addressing the need of fine-grained personalized task matching. In this paper, we argue that it is essential to match tasks to users based on a careful characterization of both the users' preference and reliability. To that end, we propose a personalized task recommender system for mobile crowdsensing, which recommends tasks to users based on a recommendation score that jointly takes each user's preference and reliability into consideration. We first present a hybrid preference metric to characterize users' preference by exploiting their implicit feedback. Then, to profile users' reliability levels, we formalize the problem as a semi-supervised learning model, and propose an efficient block coordinate descent algorithm to solve the problem. For some tasks that lack users' historical information, we further propose a matrix factorization method to infer the users' reliability levels on those tasks. We conduct extensive experiments to evaluate the performance of our system, and the evaluation results demonstrate that our system can achieve superior performance to the benchmarks in both user profiling and personalized task recommendation.

Spatial crowdsourcing, a human-centric compelling paradigm in performing spatial tasks, has drawn rising attention. Task assignment is of paramount importance in spatial crowdsourcing. Existing studies often use heuristics of various kinds to solve task assignment problems. These schemes usually only apply some specific cases, once the environment changes, the efficiency of the algorithms is significantly reduced. In this paper, we first introduce a taxonomy of task assignment in spatial crowdsourcing. Next, we design an approximation algorithm and get an efficient solution for the important problem, namely, Bounded and Heterogeneous Task Assignment (BHTA), such that the sum of the rewards of workers is maximized subject to multiple constraints. We prove that the BHTA problem is NP-hard. Subsequently, we propose a constant-ratio approximation algorithm based on partition and shifting method to achieve the assignment solution. To meet with the workers' dynamism, we further devise a greedy algorithm and provide theoretical guarantee. Experiments on synthetic and real datasets demonstrate the efficiency of our strategy over previous methods. So far as we know, this paper is the first attempt to give a constant-ratio approximation for such task assignment problems in spatial crowdsourcing.

This paper studies the problem of cHarging tAskScheduling for direcTional wireless chargEr networks (HASTE), i.e., given a set of rotatable directional wireless chargers on a 2D area and a series of offline (online) charging tasks, scheduling the orientations of all the chargers with time in a centralized offline (distributed online) fashion to maximize the overall charging utility for all the tasks. We prove that HASTE is NP-hard. Then, we prove that a relaxed version of HASTE falls within the realm of maximizing a submodular function subject to a partition matroid constraint, and propose a centralized offline algorithm that achieves$(1-\rho)(1-\frac{1}{e})$approximation ratio to address HASTE where$\rho$is the switching delay of chargers. Further, we propose a distributed online algorithm and prove it achieves$\frac{1}{2}(1-\rho)(1-\frac{1}{e})$competitive ratio. We conduct simulations and field experiments on a testbed consisting of eight off-the-shelf power transmitters and 8 rechargeable sensor nodes. The results show that our distributed online algorithm achieves 92.97 percent of the optimal charging utility, and outperforms the comparison algorithms by up to 15.28 percent in terms of charging utility.

Due to the sudden movement during the camera shoot, the videos retrieved from the hand-held mobile devices often suffer from undesired frame jitters, leading to the loss of video quality. In this paper, we present a video stabilization solution in mobile devices via inertial-visual state tracking. Specifically, during the video shoot, we use the gyroscope to estimate therotationof camera, and use the structure-from-motion among the image frames to estimate thetranslationof camera. We build a camera projection model by considering the rotation and translation of the camera, and the camera motion model to depict the relationship between the inertial-visual state and the camera's 3D motion. By fusing the inertial measurement (IMU)-based method and the computer vision (CV)-based method, our solution is robust to the fast movement and violent jitters, moreover, it greatly reduces the computation overhead in video stabilization. In comparison to the IMU-based solution, our solution can estimate the translation in a more accurate manner, since we use the feature point pairs in adjacent image frames, rather than the error-prone accelerometers, to estimate the translation. In comparison to the CV-based solution, our solution can estimate the translation with less number of feature point pairs, since the number of undetermined degrees of freedom in the 3D motion directly reduces from 6 to 3. We implemented a prototype system on smart glasses and smart phones, and evaluated the performance under real scenarios, i.e., the human subjects used mobile devices to shoot videos while they were walking, climbing or riding. The experiment results show that our solution achieves 32 percent better performance than the state-of-art solutions in regard to video stabilization. Moreover, the average processing time latency is 32.6ms, which is lower than the conventional inter-frame time interval, i.e., 33ms, and thus meets the real-time requirement for online processing.

In this article, we first study the problem of Connected wIReless Charger pLacEment (CIRCLE). That is, given a fixed number of directional wireless chargers and candidate positions, determining the placement position and orientation angle for each charger under connectivity constraint for wireless chargers such that the overall charging utility is maximized. To address CIRCLE problem, we first consider a relaxed version of CIRCLE (CIRCLE-R for short). We prove that CIRCLE-R falls into the realm of maximizing a submodular set function subject to a connectivity constraint, and propose an algorithm whose approximation ratio is at least 1.5 times better than that of the state-of-the-art algorithm. Next, we reduce the solution space for CIRCLE from infinite to finite, and propose an algorithm with a constant approximation ratio to address CIRCLE. Besides, we consider a variant of CIRCLE, CIRCLE-NB, and propose an approximation algorithm to address it. We conduct both simulation experiments and field experiments to verify our theoretical findings. The results show that our algorithm can outperform comparison algorithms by 83.35 percent.

As the Internet of Things thrives, monitors and cameras produce tons of image data every day. To efficiently process these images, many compressed imaging frameworks are proposed. A compressed imaging framework comprises two parts, image signal measurement and reconstruction. Although a plethora of measurement devices have been designed, the development of the reconstruction is relatively lagging behind. Nowadays, most of existing reconstruction algorithms in compressed imaging are optimization problem solvers based on specific priors. The computation burdens of these optimization algorithms are enormous and the solutions are usually local optimums. Meanwhile, it is inconvenient to deploy these algorithms on cloud, which hinders the popularization of compressed imaging. In this article, we dive deep into the random projection to build reconstruction algorithms for compressed imaging. We first fully utilize the information in the measurement procedure and propose a combinatorial sparse random projection (SRP) reconstruction algorithm. Then, we generalize the SRP to a novel distributed algorithm called Cloud-SRP (CSRP), which enables efficient reconstruction on cloud. Moreover, we explore the combination of SRP with conventional optimization reconstruction algorithms and propose the Iterative-SRP (ISRP), which converges to a guaranteed fixed point. With minor modifications on the naive optimization algorithms, the ISRP yields better reconstructions. Experiments on real ghost imaging reconstruction reveal that our algorithms are effective. And simulation experiments show their advantages over the classical algorithms.

Wireless Power Transfer has become a commercially viable technology to charge devices because of the convenience of no power wiring and the reliability of continuous power supply. This paper concerns the fundamental issue of wireless charger placement with electromagnetic radiation (EMR) safety. Although there are a few wireless charging schemes consider EMR safety, none of them addresses the charger placement issue. In this paper, we propose PESA, a wireless charger Placement scheme that guarantees EMR SAfety for every location on the plane. First, we discretize the whole charging area and formulate the problem into the Multidimensional 0/1 Knapsack (MDK) problem. Second, we propose a fast approximation algorithm to the MDK problem. Third, we propose a near optimal scheme to improve speed by double partitioning the area. We prove that the output of our algorithm is better than $(1-\epsilon)$ of the optimal solution to PESA with a smaller EMR threshold $(1-\epsilon /2)R_{t}$ and a larger EMR coverage radius $(1+\epsilon /2)D$ . We conducted both simulations and field experiments to evaluate the performance of our scheme. Our experimental results show that in terms of charging utility, our algorithm outperforms the comparison algorithms.

In networks where massive sources make observations of same entities, we intend to seek thetruth– the most trustworthy value of each entity from conflicting information claimed by multiple sources. Various methods are proposed for accurately inferring both source reliability and truths, yet relying heavily on centralized settings that incur tremendous overhead to source side. In this paper, we offer adecentralizeddesign of truth discovery task that can fit favorably to the environments with limited resources. Considering that sources forming the connected network and making individual observations, we undertake the joint maximum likelihood estimation (MLE) of truth and source reliability. Our decentralization framework simply allows each source to maintain local information exchange at a time, and computes very basic functions of data observations. To this end, we facilitate the decentralization by simplifying the MLE problem into optimizing an objective function. Upon the proof of NP-hardness, two proposed decentralized algorithms (exact and approximation) are decentralized and randomized via a combination of algorithms from their centralized counterparts that ensure performance guarantee. The derived time complexity features explicit data/network dependent terms, which leads to further acceleration in truth finding. Remarkably, in two well connected networks like random geometric and preferential attachment graphs, the accelerated approximation method enjoys logarithmic time complexity while preserving comparable accuracy to the centralized counterparts. The effectiveness of the proposed decentralizations are further empirically confirmed.

Due to the distributed and asynchronous nature in data plane, the packets can be forwarded into a loop during routing updates. Software-Defined Networks (SDNs) enable a controller to schedule the update operations of routing rules in a loop-free manner. However, the existing rule replacement mechanism cannot give an efficient solution for large-scale update scenarios. In this paper, we make use of the ingress port field in forwarding rules to design a loop-free update scheme, Inport-Matching Update (IMU), for per-flow and destination-based unicast, which significantly improves efficiency of solving update scenarios and reduces update rounds. In both our proposed and previous mechanisms, a subproblem called Rule Replacement Scheduling (RRS) problem needs to be solved and we prove that the scale of RRS problem can be reduced at least by half. Moreover, we use the inclusion-exclusion principle to rigorously prove that the probability that an RRS problem under IMU has a trivial solution, i.e., one round, is always higher than 83.7%, given a random per-flow unicast update scenario with ingress port rules. Experimental results show that our proposed Inport-Matching Update mechanism remarkably reduces the average number of rounds and runtime for solving the RRS problem compared to the existing rule replacement mechanism.

Due to the benefits of small latency, low energy consumption and increased data rate, device-to-device (D2D) communication is recognized as one of the promising techniques in the 5G era. However, the distributed nature of D2D communication makes it non-trivial to generate symmetric keys for the involving parties. Many efforts have been devoted to dynamically generate cryptographic keys for D2D communication in mobile network. However, most of them have limited applicability to practical scenarios due to low key generation efficiency or limited compatibility with commercial mobile devices. In this paper, we design an ambient light based key generation approach, which works on commercial off-the-shelf mobile devices, and achieves high key generation efficiency. We observe that mobile devices (e.g., smartphones, tablets) are often equipped with ambient light sensors and devices sense different light intensities at different angles towards the light source. We conduct a set of measurement study, and the results show that the light sensor data have several nice properties, i.e., space-varying, sensitive to the angle of measuring device, and time varying. These properties demonstrate that the ambient light is a good medium for secret key generation. We implement a prototype on different mobile devices, and conduct extensive experiments to evaluate its performance. The experiment results show that compared with the state-of-the-art key generation methods, our approach has a superior performance in key generation efficiency and robustness.

Bike sharing systems have been widely deployed around the world in recent years. A core problem in such systems is to reposition the bikes so that the distribution of bike supply is reshaped to better match the dynamic bike demand. When the bike-sharing company or platform is able to predict the revenue of each reposition task based on historic data, an additional constraint is to cap the payment for each task below its predicted revenue. In this paper, we propose an incentive mechanism called TruPreTar to incentivize users to park bicycles at locations desired by the platform toward rebalancing supply and demand. TruPreTar possesses four important economic and computational properties such as truthfulness and budget feasibility. Furthermore, we prove that even when the payment budget is tight, the total revenue still exceeds or equals the budget. Otherwise, TruPreTar achieves 2-approximation as compared to the optimal (revenue-maximizing) solution, which is close to the lower bound of at least √2 that we also prove. Using an industrial dataset obtained from a large bike-sharing company, our experiments show that TruPreTar is effective in rebalancing bike supply and demand and, as a result, generates high revenue that outperforms several benchmark mechanisms.

In this paper, we consider the problem of designing a novel ATTack scheme under charger cApture attaCK(ATTACK) in wireless rechargeable sensor networks. That is, there are a number of wireless chargers with directional antenna and rechargeable directional devices deployed in the 2D plane. An intelligent adversary selects a limited number of chargers to be captured and adjust the compromised chargers' power factor such that the overall attacking utility of devices is maximized. To solve our ATTACK problem, we investigate an attacking algorithm with a constant approximation ratio with lightweight timing complexity. We conduct simulation experiments to verify the performance of our proposed attacking algorithm. The results shows that our proposed attacking algorithm can outperform the comparison algorithm by 69.95%.

In this paper, we consider the problem of area charging, that is, assuming that there is a mobile charger (MC) equipped with a directional wireless charger whose charging area is in the shape of a sector, and the MC can only recharge sensors on the boundary of an Area of Interest (AOI), how to design an efficient charging scheme for the MC to recharge any sensor inside the AOI with efficient energy while its overall charging time is minimized. To address this problem, we first partition the AOI into two types of subareas, i.e., rectangle-like subareas and sector-like subareas, based on the Medial Axis of the AOI. Second, we propose rectangle-based moving strategy for the rectangle-like subareas, and sector-based rotating strategy for the sector-like subareas, respectively, for the MC. Finally, we calculate charging time of all subareas based on the above two moving strategies, and prove that our algorithm achieves a constant approximation ratio. We evaluate our algorithm by conducting extensive simulation. The results show that on average, other comparison algorithms require at least 18:70 times of charging time of that of our algorithm.

Although the logically-centralized perspective is offered in Software-Defined Networks (SDNs), the data plane is still distributed in nature. Update commands sent by the centralized controller are executed asynchronously and independently in each switch. The timed SDNs enable synchronous and coordinate update operations as each update command can be triggered by a pre-defined time point. Prior work on timed update mainly focuses on how to produce a congestion-free update sequence, whereas finding a congestion-free timed update sequence may be too long to be applied in practice, even worse, such an update order may not exist. In this paper, we propose Chronus+, a timed update system that utilizes switch buffer to shorten the update time while minimizing the switch buffer size during updates. We formulate the Minimum Switch Buffer Size Problem (MSBSP) as an optimization program and show its hardness. A set of efficient algorithms is proposed to determine a timed update sequence in polynomial time. Extensive evaluations in Mininet and large-scale simulations show that Chronus+can reduce the update time by at least 17% and the switch buffer by at least 27% compared with state-of-the-art approaches.

The society's insatiable appetites for personal data are driving the emergency of data markets, allowing data consumers to launch customized queries over the datasets collected by a data broker from data owners. In this paper, we study how the data broker can maximize her cumulative revenue by posting reasonable prices for sequential queries. We thus propose a contextual dynamic pricing mechanism with the reserve price constraint, which features the properties of ellipsoid for efficient online optimization, and can support linear and non-linear market value models with uncertainty. In particular, under low uncertainty, our pricing mechanism provides a worst-case regret logarithmic in the number of queries. We further extend to other similar application scenarios, including hospitality service and online advertising, and extensively evaluate all three application instances over MovieLens 20M dataset, Airbnb listings in U.S. major cities, and Avazu mobile ad click dataset, respectively. The analysis and evaluation results reveal that our proposed pricing mechanism incurs low practical regret, online latency, and memory overhead, and also demonstrate that the existence of reserve price can mitigate the cold-start problem in a posted price mechanism, and thus can reduce the cumulative regret.

With the rapid development of mobile network and devices, spatial crowdsourcing (SC) has recently attracted much attention. For the improvement of quality of service (QoS) in spatial crowdsourcing platforms, existing works usually adopt the many-to-one strategy - assigning multiple workers as a team for each published task. However, such an allocation scheme fails to consider team characteristics which can strongly affect the QoS for some experience-sensitive collaborative tasks. In this paper, we jointly consider two team characteristics to further improve the QoS: Diversity, which is the union of experiences within a team and Affinity, which is how efficiently team members collaborate. Inspired by these two characteristics, we study an important problem, namely, Affinitive Diversity-Aware Spatial Crowdsourcing (ADA-SC), which aims to find an allocation scheme, such that each team satisfies the affinity requirement of the corresponding task and maximizes the team diversity under budget and spatial constraints. Since ADA-SC is proven to be NP-hard by reduction from the set cover problem with nonlinear constraints, we propose two submodular approximation algorithms with pruning strategies for two single-task scenarios. Then a greedy-based algorithm is designed for the multi-task scenario. Extensive experiments on real and synthetic data verify the effectiveness of our proposed methods.

Spatial crowdsourcing has a high demand on timeliness, it requires workers to finish tasks in a proper order, and is reflected by the exponentially reduced utility. Fairness among workers is an important issue to the attraction of the platform. In this paper, we propose Utility-Fairness Index (UFI) to evaluate the performance of the crowdsourcing platform, based on which we address the Fairness-Aware Task Planning (FATP) problem that aims at task planning, such that the utility is maximized while fairness is well maintained. It is different from the previous studies concerning task assignment in which the completion order of the task and the fairness among workers is often ignored. Before digging deeper into the formulated FATP problem, we first prove the NP-hardness of its offline version, then we prove that there is no online algorithm with a constant competitive ratio for it. Two algorithms, Traversal Search (TS) and Advanced Nearest Neighbor (ANN), are proposed to solve this problem. To optimize the planning, we adopt Gaussian Mixture Model (GMM) to reveal the distribution regularity of tasks in these algorithms. Experiments on both synthetic and real datasets validate the efficiency and effectiveness of our algorithms.

Web based social services like Twitter and Weibo are popular nowadays. A large number of users generate massive text-based information continuously and as a result, many topic popularity prediction models arise. However, most of them simply use forwarding or view count to measure popularity, ignoring semantic relations between posts and the topic. Moreover, few model use the important component of opinion - sentiment, to enhance topic popularity prediction. Therefore, in this paper, we propose a attention-based convolutional network model to generate sentiment-aware time series and predict topic popularity. Our model quantifies popularity more accurately with posts' semantic information and incorporates sentiment into popularity prediction. Concretely, our model is composed of a semantic-aware popularity quantification metric, a sentiment intensity detection neural network, and an Autoregressive convolutional network prediction scheme. We have conducted extensive experiments to prove that our model outperforms the existing popularity prediction models on a real-world Twitter dataset. We find that sentiment time series have the ability to improve popularity prediction accuracy significantly. Furthermore, we find our prediction scheme is superior to other models in utilizing sentiment information.

Long Range (LoRa) communication, thanks to its wide network coverage and low energy operation, has attracted extensive attentions from both academia and industry. However, existing LoRa-based Wide Area Network (LoRaWAN) suffers from severe inter-network interference, due to the following two reasons. First, the densely-deployed LoRa ends usually share the same network configurations, such as spreading factor (SF), bandwidth (BW) and carrier frequency (CF), causing interference when operating in the vicinity. Second, LoRa is tailored for low-power devices, which excludes LoRaWAN from using the listen-before-talk (LBT) mechanisms commonly used in wireless communication technologies, such as WiFi and ZigBee -LoRaWAN has to use the duty-cycled medium access policy and thus being incapable of channel sensing or collision avoidance. To mitigate the inter-network interference, we propose a novel solution achieving the online concurrent transmissions at LoRa gateway, called OCT, which recovers collided packets at the gateway and thus improves LoRaWAN's throughput. Moreover, OCT achieves the online concurrent transmission using only LoRa's (de)modulation information, thus can be easily deployed at LoRa gateway. We have implemented and evaluated OCT on USRP platform and commodity LoRa ends, showing OCT achieves: (i) >90% packet reception rate (PRR), (ii) 3 × 10-3bit error rate (BER), (iii) 2x and 3x throughput in the scenarios of two- and three- packet collisions respectively, and (iv) reducing 67% latency compared with state-of-the-art.

In this work, we study the problem of Robust Server Placement (RSP) for edge computing, i.e., in the presence of uncertain edge server failures, how to determine a server placement strategy to maximize the expected overall workload that can be served by edge servers. We mathematically formulate the RSP problem in the form of robust max-min optimization, derived from two consequentially equivalent transformations of the problem that does not consider robustness and followed by a robust conversion. RSP is challenging to solve, because the explicit expression of the objective function in RSP is hard to obtain, and RSP is a robust max-min problem with a matroid constraint and a knapsack constraint, which is still an unexplored problem in the literature. To address the above challenges, we first investigate the special properties of the problem, and reveal that the objective function is monotone submodular. We then prove that the involved constraints form a p-independence system constraint, where p is a constant value related to the ratio of the coefficients in the knapsack constraint. Finally, we propose an algorithm that achieves a provable constant approximation ratio in polynomial time. Both synthetic and trace-driven simulation results show that, given any maximum number of server failures, our proposed algorithm outperforms three state-of-the-art algorithms and approaches the optimal solution, which applies exhaustive exponential searches.

A distributed control plane is more scalable and robust in software defined networking. This paper focuses on controller load balancing using packet-in request redirection, that is, given the instantaneous state of the system, determining whether to redirect packet-in requests for each switch, such that the overall control plane response time (CPRT) is minimized. To address the above problem, we propose a framework based on Lyapunov optimization. First, we use the drift-plus-penalty algorithm to combine CPRT minimization problem with controller capacity constraints, and further derive a non-linear program, whose optimal solution is obtained with brute force using standard linearization techniques. Second, we present a greedy strategy to efficiently obtain a solution with a bounded approximation ratio. Third, we reformulate the program as a problem of maximizing a non-monotone submodular function subject to matroid constraints. We implement a controller proto-type for packet-in request redirection, and conduct trace-driven simulations to validate our theoretical results. The results show that our algorithms can reduce the average CPRT by 81.6% compared to static controller-switch assignment, and achieve a 3× improvement in maximum controller capacity violation ratio.

As the most widely used outdoor navigation system, GPS can provide accurate localization on the horizontal plane. However, the vertical localization accuracy exhibits a poor performance. Vehicles on the elevated road usually receive wrong navigation instructions since GPS fails to detect vehicles' vertical location. In this paper, we present a vertical location system named VLD for vehicles in metropolises mainly leveraging smartphones' barometers, a low-power sensor found in an increasing number of smart devices. When initially on the ground, VLD combines the height and angle detection algorithms to confirm the vertical location with low complexity. Once vehicles have exited the ground and start to travel on the elevated road, a pressure-height model trained by a novel proposed sensor fusion algorithm is activated to measure vehicles' relative height. Then, it tracks the relative height in real time according to a pressure-temperature model which is calibrated by current weather. Comparing the relative height with the single-level elevated road's height, VLD can determine which level vehicles travel on in many highway interchanges and when they exit the elevated road. Experiments covering one month demonstrate that the detection accuracy of VLD exceeds 99% under different weather conditions, and it shows a more accurate relative height measurement compared to available literature, including Baidu Maps and one barometer based application-Altitude. Finally, VLD demonstrates a high efficiency with respect to both the detection delay and power consumption.

The development of Wireless Power Transfer (WPT) facilitates wireless rechargeable sensor networks (WRSNs) receiving considerable attention in the sensor network research community. Most existing works mainly focus on general charging patterns and metrics while overlooking the precedence constraints among tasks, resulting in charging inefficiency. In this paper, we are the first to advance the issue of Scheduling wireless Charging tasks with Precedence Constraints (SCPC) with the optimization objective of minimizing the completion time of all the charging tasks under precedence constraints while guaranteeing that the energy capacity of mobile charger (MC) is not exhausted, and the deadlines of charging tasks are not exceeded. In order to address the problem, we first propose the priority-based topological sort scheme to derive an unique feasible sequence on directed acyclic graph (DAG). Then, we combine the proposed priority-based topological sort scheme with the procedure of genetic algorithm to obtain the final solution through a series of genetic operators. Finally, we conduct extensive simulations to validate our proposed algorithm under the condition of three different network sizes. The results show that our proposed algorithm outperforms the other two comparison algorithms by up to 35.5% in terms of completion time.

A large-scale image collection is crucial to the success of 3D reconstruction. Crowdsourcing, as a new pattern, can be utilized to collect high-quality images in an efficient way. However, the sensitive information in images may be exposed during the image transmission process. The general privacy policies perhaps will cause the loss or change of critical information, which may give rise to a decline in the performance of 3D reconstruction. Hence, how to achieve image privacy-preserving while guaranteeing to reconstruct a complete 3D model is important and significant. In this paper, we propose PicPrivacy to address this problem, which consists of three parts. (1) Using a pre-trained deep convolution neural network to segment sensitive information and erase it from images. (2) Using a GAN-based image feature completion algorithm to repair blank regions and minimize the absolute information gap between generated images and raw ones. (3) Taking generated images as the input of 3D reconstruction and using a structure-from-motion algorithm to reconstruct 3D models. Finally, we extensively evaluate the performance of PicPrivacy on realworld datasets. The results demonstrate that PicPrivacy not only achieves individual privacy-preserving but also can guarantee to create complete 3D models.

Group activities like group dance, military parade, or radio gymnastics have excellent ornamental value with its grand scale and uniform movements, while it also introduces difficulties in practice for coaches to monitor and correct the movements and locations of each participator. Wireless body area network (WBAN) is a promising direction for accurate motion tracking in large-scale group activities. Light-weight sensors can be deployed on participant bodies and transmit motion and channel sensing data through wireless spectrum for further analysis. To keep the high quality of service (QoS) of the WBAN on group activity monitoring and correction, three aspects should be achieved by careful designing: (i) less energy consumption; (ii) higher analysis accuracy; and (iii) lower feedback latency. Due to the low-rankness of motion and channel sensory data, we propose GroupCoach, a Compressed Sensing (CS) based group activity monitoring and correction system. The data is collected and reconstructed by CS, where the spatial and temporal stabilities of group activities are explored for a higher reconstruction accuracy. These reconstructed sensory data are further compared with their anchor values for faulty movement detection and correction. The channel attenuation caused by body shielding is reduced by a near-to-far diffusion model with the consideration of the body impact factor. The correction suggestions are finally fed back to sensors for guidance. Evaluations based on the prototype deployed on real group activity participators prove the high QoS of the GroupCoach. It achieves low sensor energy consumption, high data reconstruction accuracy, accurate faulty motion detection and correction, together with fast alert.

Compared with hardware implementation, the software packet scheduler uses the packet queuing data structure and a ranking function according to different dimensions to flexibly determine the packet dequeue order, which can significantly shorten the renewal cycles and increase the function deployment flexibility. The key data structure in prior work either bounds the number of rank or suffers from high computation overhead. In addition, they only support a single dimension and do not scale well. In this paper, we present Proteus, a software packet scheduling system that supports multi-dimensional and arbitrary numbers of ranks. We design a k-dimension heap data structure and develop “push” and “pop” algorithms to perform “enqueue” and “dequeue” operations. Furthermore, we implement a prototype of Proteus in software switch. Extensive experiments on BESS and numerical simulations show that Proteus can decrease the computation overhead, save the storage space and run much faster than state of the art.