Engaging applications with diverse SLO requirements has become indispensable for production-scale LLM serving systems. However, existing systems rely on iteration-level scheduling, which enforces inflexible, unified execution across multi-SLO workloads, significantly constraining the serving efficiency.

The surge in low-altitude economic activities has spurred a significant interest in sensing Unmanned Aerial Vehicles (UAVs). With the widespread deployment of 5G infrastructure and the increasing prominence of integrated sensing and communication, monitoring UAVs via 5G base stations is a natural consideration. However, the rapid Doppler shifts of UAVs and sparse 5G reference signals violate Nyquist sampling requirements. To bridge this gap, we propose LaSen, which merges reference and downlink data signals for sensing. A key challenge stems from the fact that the combination of the periodic reference signals and stochastic data signals constitutes a non-uniform, time-varying measurement matrix. LaSen formulates the tracking of UAVs as a sparse recovery problem, where the target’s kinematics are reconstructed from non-uniform, sub-Nyquist observations. LaSen overcomes the challenges of volatile 5G signal patterns in an iterative way, starting from good measurements as an anchor and progressively refining the suboptimal measurements. Real-world experiments show that LaSen significantly extends the velocity sensing capability, where the measurable speed is up to 20.2 m/s. LaSen can detect drones at a distance of 108 m and can continuously track the distance and velocity of multiple targets, even when the downlink channel is sparsely and dynamically occupied. This work demonstrates the feasibility of high-speed target sensing in next-generation dual-function 5G/6G infrastructures.

Autonomous navigation for Unmanned Aerial Vehicles (UAVs) presents significant challenges due to the limited onboard computational resources, which often restrict deployed deep neural networks to shallow architectures incapable of handling complex environments. Additionally, offloading tasks to remote edge servers introduces high latency, creating an inherent trade-off in system design. To address these limitations, we propose CoDrone—the first cloud-edge-end collaborative computing framework that integrates foundation models into autonomous UAV cruising scenarios—effectively leveraging foundation models to enhance the performance of resource-constrained unmanned aerial vehicle platforms. To reduce both onboard computation and data transmission overhead, CoDrone employs grayscale imagery for the navigation model. When enhanced environmental perception is required, CoDrone leverages the edge-assisted foundation model Depth Anything V2 for depth estimation and introduces a novel, one-dimensional occupancy grid–based navigation method—enabling fine-grained scene understanding while significantly advancing the efficiency and representational simplicity of autonomous navigation. A key component of CoDrone is a Deep Reinforcement Learning (DRL)-based neural scheduler that seamlessly integrates depth estimation with autonomous navigation decisions, enabling real-time adaptation to dynamic environments. Furthermore, the framework introduces a UAV-specific vision language interaction module, which incorporates domain-tailored low-level flight primitives to enable effective interaction between the cloud foundation model, the Vision Language model, and the UAV. The introduction of VLM enhances open-set reasoning capabilities in complex and previously unseen scenarios. We implement a prototype of CoDrone and conduct extensive evaluations in the AirSim simulation environment. Experimental results demonstrate that CoDrone significantly outperforms baseline methods under varying flight speeds and network conditions, achieving a 40% increase in average flight distance and a 5% improvement in average Quality of Navigation.

Video analytics (VA) has become essential in applications, yet it poses significant challenges related to privacy preservation, network bandwidth, and computational resources. With the increasing deployment of high-definition cameras, privacy concerns and resource constraints are becoming critical barriers to the widespread adoption of VA systems. Existing privacy-preserving techniques are often static, inefficient, and fail to adapt to dynamic, real-time scenarios. In this paper, we propose Parachute, a dynamic, resource-aware and privacy-preserving video analytics system that adaptively switches between a local mode and a collaborative mode in response to traffic conditions. The system uses local reinforcement learning to enable each individual camera to operate independently, and switches to multi-agent reinforcement learning for coordinated optimization when local resources become limited. Experiments on real-world datasets demonstrate that Parachute effectively balances detection accuracy and privacy protection, outperforming baseline methods under bandwidth constraints.

How to preserve the data privacy during the training of deep neural network (DNN) is a key security concern in the artificial intelligence era. However, most existing solutions based on homomorphic encryption and Trusted Execution Environment (TEE) are incompatible with heterogeneous Neural Network Accelerators (NNAs), leading to significant performance loss. We propose a novel method based on vector decomposition to allocate operators across different NNAs, ensuring both throughput and privacy simultaneously. Furthermore, based on this approach, we have designed a compiler that automatically converts front-end model descriptions into backend encrypted computation graphs, which is running securely over trusted and untrusted hardware. This compiler heuristically determines the allocation scheme based on hardware affinity and cross-hardware communication costs, significantly reducing additional overhead. Experimental results demonstrate that our method does not incur extra accuracy costs and achieves a throughput significantly higher than existing methods. Deploying our approach at scale on a platform with a billion users, we have verified its negligible impact on real-world operations while ensuring the privacy protection capability for cross-domain data.

Although Federated Learning (FL) can enable advanced autonomous driving via leveraging massive distributed data in vehicular networks, vehicle mobility causes frequent connection interruptions, hindering the FL process. In this paper, we propose a novelMobility-AwareVehicularFL(MAVFL) scheme, which can accelerate the training process in dynamic vehicular networks via adaptive vehicle selection and sparse training. Specifically, the MAVFL dynamically selects participating vehicles based on their locations and training loss. By incorporating adaptive model sparsification, the proposed scheme dynamically proceeds with sparse masks during vehicle local training, thereby reducing communication overhead while preserving model accuracy. We conduct a rigorous convergence analysis to uncover how vehicle mobility and model sparsification affect convergence rate. Furthermore, we formulate an optimization problem to accelerate the training process, which jointly optimizes vehicle selection, sparsification ratio, and bandwidth allocation to minimize training delay. To solve the problem, we employ the Lyapunov optimization method to decouple the long-term problem into a series of instantaneous subproblems. Next, a generalized Benders decomposition method structures the original problem into a master subproblem for vehicle selection and a primal subproblem for bandwidth allocation and sparsification ratio selection. The optimal solutions are derived via alternating iterations between these problems. Extensive simulation results based on the SUMO simulator demonstrate that the MAVFL accelerates model convergence by up to 14% and reduces communication overhead by up to 26% while preserving model accuracy, as compared to the state-of-the-art benchmarks.

Multi-party Mobile Virtual Reality (MMVR) enables multiple mobile users to share virtual scenes for an immersive multimedia experience in scenarios such as gaming, social interaction, and industrial mission collaboration. Dynamic 3D Point Cloud (DPCL) is an emerging representation form of MMVR that can be consumed as a free-viewpoint video with 6 degrees of freedom. With limited on-device resources, it is a challenge to achieve a satisfying frame rate for DPCL rendering, which makes edge-assisted rendering a practical solution. However, repeated loading of DPCL scenes with a substantial amount of metadata introduces a significant redundancy overhead that cannot be overlooked when enabling multiple edge servers to support the rendering requirements of user groups. In this paper, we design PoClVR, an edge-assisted DPCL rendering system for MMVR applications, which introduces an object-level splitting mode to alleviate performance bottlenecks caused by redundant loading. In addition, PoClVR dynamically selects the splitting mode and scheduling decisions to adapt to varying task requirements and available computational resources, thereby improving overall system efficiency. To evaluate the performance of PoClVR, we implement and deploy a realistic prototype system and also conduct large-scale trace-driven simulations. The experimental results show that PoClVR can reduce resource usage by up to approximately 49.3% under different task requirements and resource conditions, while decreasing bottleneck performance degradation by up to 77.3%.

As large language models (LLMs) increasingly integrate into every aspect of our work and daily lives, there are growing concerns about user privacy, which push the trend toward local deployment of these models. There are a number of lightweight LLMs (e.g., Gemini Nano, LLAMA2 7B) that can run locally on smartphones, providing users with greater control over their personal data. As a rapidly emerging application, we are concerned about their performance on commercialoff- the-shelf mobile devices. To fully understand the current landscape of LLM deployment on mobile platforms, we conduct a comprehensive measurement study on mobile devices. While user experience is the primary concern for endusers, developers focus more on the underlying implementations. Therefore, we evaluate both user-centric metrics-such as token throughput, latency, and response quality-and developer-critical factors, including resource utilization, OS strategies, battery consumption, and launch time. We also provide comprehensive comparisons across the mobile system-on-chips (SoCs) from major vendors, highlighting their performance differences in handling LLM workloads, which may help developers identify and address bottlenecks for mobile LLM applications. We hope that this study can provide insights for both the development of on-device LLMs and the design for future mobile system architecture.

The rapid proliferation of edge computing infrastructure has significantly accelerated the adoption of edge-native applications, ranging from autonomous vehicles to augmented reality and real-time analytics. Microservice, renowned for its lightweight, loosely coupled, and modular architecture, has emerged as the de-facto standard for developing edge native applications. However, the resource scarcity and heterogeneity, coupled with request dynamics in edge environments, pose substantial challenges for effective microservice orchestration. To address these challenges, we propose MicroEdge, an online optimization framework designed for cost-efficient microservice orchestration in edge-native environments. MicroEdge employs a multi-level optimization approach by strategically coordinating four key dimensions: microservice placement, layer placement, layer pulling, and user request scheduling. The framework confronts two fundamental challenges in solving this joint optimization problem: (1) the time-coupled nature of long-term holistic cost minimization, and (2) the NP-hardness of the underlying problem. MicroEdge tackles these dual challenges by integrating a regularization method for online algorithm design and a dependent rounding technique for approximation algorithm design. Both rigorous theoretical analysis and extensive simulations driven by realistic Alibaba microservice workload traces validate the efficacy of MicroEdge.

Today's video streaming service providers have exploited cloud-edge collaborative networks for video delivery across geo-distributed edge clusters and end users. The existing content delivery network (CDN) scheduling and adaptive bitrate algorithms may not fully utilize edge resources or lack a global control to optimize resource sharing. The emerging super-resolution (SR) approach can unleash the potential of leveraging computation resources to compensate for bandwidth consumption, by producing high-quality videos from low-resolution contents. Yet the uncertain SR resource sensitivity and its interplay with bitrate adaptation are under-explored. In this work, we proposeRosevin, the first resource scheduler that jointly decides the bitrates and fine-grained resource allocation to perform SR at the edge, which can learn to optimize the long-term QoE for distributed end users. To handle the time-varying and complex space of decisions as well as a non-smooth objective function,Rosevinrealizes a novel online combinatorial learning algorithm, which nicely integrates convex optimization theories and online learning techniques, addressing the switching cost issues. In addition to theoretically analyzing its performance, we implement an SR-assisted video streaming prototype ofRosevinand demonstrate its advantages over several video delivery benchmarks.

Advances in artificial intelligence (AI) including foundation models (FMs), are increasingly transforming human society, with smart city driving the evolution of urban living. Meanwhile, vehicle crowdsensing (VCS) has emerged as a key enabler, leveraging vehicles' mobility and sensor-equipped capabilities. In particular, ride-hailing vehicles can effectively facilitate flexible data collection and contribute towards urban intelligence, despite resource limitations. Therefore, this work explores a promising scenario, where edge-assisted vehicles perform joint tasks of order serving and the emerging foundation model finetuning using various urban data. However, integrating the VCS AI task with the conventional order serving task is challenging, due to their inconsistent spatio-temporal characteristics: (i) The distributions of ride orders and data point-of-interests (PoIs) may not coincide in geography, both following a priori unknown patterns; (ii) they have distinct forms of temporal effects, i.e., prolonged waiting makes orders become instantly invalid while data with increased staleness gradually reduces its utility for model fine-tuning. To overcome these obstacles, we propose an online framework based on multi-agent reinforcement learning (MARL) with careful augmentation. A new quality-of-service (QoS) metric is designed to characterize and balance the utility of the two joint tasks, under the effects of varying data volumes and staleness. We also integrate graph neural networks (GNNs) with MARL to enhance state representations, capturing graph-structured, time-varying dependencies among vehicles and across locations. Extensive experiments on our testbed simulator, utilizing various real-world foundation model fine-tuning tasks and the New York City Taxi ride order dataset, demonstrate the advantage of our proposed method.

In recent years, volumetric videos have gradually prospered as an intriguing video paradigm, offering users a fully immersive viewing experience with six Degrees of Freedom (DoF). However, most current live volumetric video streaming methods struggle to facilitate the real-time performance requirements due to the nature of frequent user interactions and the complexity of network environments during video playback. Inspired by the correlation between the human visual effects and adjacent frame motion features, we proposeCetus, a context-aware cross-layer coordination system for live volumetric videos. First, we present an application-layer Neural Radiance Fields (NeRF)-based codec framework that leverages spatio-temporal semantic information for optimizing the compression quality of each video frame. Second, we exploit a flexible cross-layer coordination framework that seamlessly integrates frame drop strategy with partially reliable transmission, orchestrating transport protocols and application-informed rates to enhance the Quality of Experience (QoE) for multiple users. Furthermore, we develop a lightweight branching decision tree algorithm that adaptively makes fine-grained frame drop decisions. Experimental evaluations of our implemented system prototype demonstrate that Cetus significantly outperforms existing baseline approaches. Compared to the state-of-the-art baselines, Cetus effectively improves video frame rate by at least 24.7% and video quality by an average of 32.6%.

Cloud and edge computing service providers (SPs) provide heterogeneous computing services to users, which forms the computing market. However, users’ service selections among SPs are unbalanced, resulting in inefficient resource utilization. In this paper, we analyze users’ service selection behaviors and design efficient pricing mechanisms to optimize the social welfare of the computing market. Considering the huge number of users and heterogeneous service providers, users can hardly acquire complete information to make their decisions, and we model users’ interactions as a dynamic service selection evolutionary game. Analyzing the evolutionary stable state (ESS) of the game and designing efficient pricing mechanisms for edge-cloud computing markets are challenging due to the implicit relationship between prices and users’ service selection dynamics, and the heterogeneity of service providers and user populations. We first investigate a single-population scenario where users are homogeneous, for which we prove that the ESS is unique. We design a static pricing mechanism which depends on SPs’ marginal costs and computation capacities, and a dynamic pricing mechanism which also depends on SPs’ real-time congestion tax. We prove that under our pricing mechanisms, the ESS is the socially optimal state and is globally asymptotic stable. We then analyze the general multi-population scenario where users are heterogeneous, for which we prove that the ESS exists but may be not unique. We design a static pricing mechanism (and a dynamic pricing mechanism) which depends on SPs’ marginal costs and the social-optimal average delay costs (and the real-time average delay costs). We prove that under our pricing mechanisms, the socially optimal state is an ESS and is asymptotically stable. Simulation results validate the effectiveness of our designed pricing mechanisms.

As a promising computing paradigm, edge-cloud computing network integrates the ubiquitous computing resources on the cloud and edge servers to provide high-quality services, but is susceptible to complicated distributed denial-of-service (DDoS) attacks. Although some works have studied edge DDoS mitigation, the cooperative defense among cloud and edge servers considering the DDoS attacker’s strategic attack strategy is yet to be explored. In this paper, we model the interactions between the DDoS attacker and defenders as a two-stage dynamic game and propose a distributed cooperative defense scheme. In Stage I, the DDoS attacker strategically launches different amounts of malicious traffic to different edge servers to maximize the total filtering cost. In Stage II, edge servers under attack (i.e., defenders) filter their traffic on the cloud and other edge servers cooperatively to minimize the total filtering cost. The defenders’ problem in Stage II is NP-hard and we solve the problem by modeling defenders’ behaviors as a selfish filtering game. We prove that the selfish filtering game admits a unique Nash equilibrium (NE) with guaranteed social efficiency, and design both a centralized algorithm and a distributed algorithm to calculate the NE. For the attacker’s problem in Stage I, we first analyze a special case where each edge server has the same amount of normal traffic, and design a low-complexity algorithm to calculate the optimal attack strategy. We then analyze the general case where edge servers have different amounts of normal traffic, for which we derive the approximate optimal solution. Simulation results show that the DDoS attacker tends to launch attacks to all edge servers to reduce their cooperative defense capability, and our proposed distributed cooperative defense mechanism can effectively reduce the total filtering cost compared with existing benchmark defense mechanisms.

Large language models (LLMs) have unlocked a plethora of powerful applications at the network edge, such as intelligent personal assistants. Data privacy and security concerns have prompted a shift towards edge-based fine-tuning of personal LLMs, away from cloud reliance. However, this raises issues of computational intensity and resource scarcity, hindering training efficiency and feasibility. While current studies investigate parameter-efficient fine-tuning (PEFT) techniques to mitigate resource constraints, our analysis indicates that these techniques are not sufficiently resource-efficient for edge devices. Other studies focus on exploiting the potential of edge devices through resource management optimization, yet are ultimately bottlenecked by the resource wall of individual devices. To tackle these challenges, we proposePAC+, a resource efficient collaborative edge AI framework for in-situ personal LLMs fine-tuning.PAC+breaks the resource wall of personal LLMs fine-tuning with a sophisticated algorithm-system co-design. (1) Algorithmically,PAC+implements a personal LLMs fine-tuning technique that is efficient in terms of parameters, time, and memory. It utilizes Parallel Adapters to circumvent the need for a full backward pass through the LLM backbone. Additionally, an activation cache mechanism further streamlining the process by negating the necessity for repeated forward passes across multiple epochs. (2) Systematically,PAC+leverages edge devices in close proximity, pooling them as a collective resource for in-situ personal LLMs fine-tuning, utilizing a hybrid data and pipeline parallelism to orchestrate distributed training. The use of the activation cache eliminates the need for forward pass through the LLM backbone, enabling exclusive fine-tuning of the Parallel Adapters using data parallelism. Extensive evaluation of the prototype implementation demonstrates thatPAC+significantly outperforms existing collaborative edge training systems, achieving up to a$9.7\times$end-to-end speedup. Furthermore, compared to mainstream LLM fine-tuning algorithms,PAC+reduces memory footprint by up to$88.16\%$.

A user with limited computation resources can address his delay-sensitive and computation-intensive tasks through task offloading to nearby edge servers, by purchasing both network and computation resources from profitseeking providers. We identify a substitutability property of computation and network resources for realizing the delay requirement. That is, to reduce task delay, the user can purchase more network resources to reduce transmission delay or more computation resources to reduce computation delay. This property significantly affects the user's purchase behavior and leads to strategic interactions between the computation service provider (CSP) and the network service provider (NSP), which have not been systematically studied yet. To this end, we formulate a two-stage Stackelberg game. In Stage I, one CSP and one NSP set their prices. In Stage II, each user decides offloading ratio and the amount of resources to purchase. By deriving the closed-form solutions in Stage II, we analytically conclude that the substitutability affects the user's decision through the network price to computation price ratio. We then incorporate the solution in Stage II into Stage I and analyze the service providers' pricing under two market structures. In the cooperative setting, where two service providers are integrated and jointly maximize their total profit, they would flexibly adjust the price ratio based on computation and network costs. In the competitive setting, where they are separate firms and aim to maximize their own profit, we formulate a pricing game and characterize a counter-intuitive equilibrium: the service providers would set high prices instead of low prices. Experimental results show that users benefit from service providers' competitive interactions.

We focus on the problem of fusing two or more heterogeneous large language models (LLMs) to leverage their complementary strengths.One of the challenges of model fusion is high computational load, specifically in fine-tuning or aligning vocabularies.To address this, we propose Cool-Fusion, a simple yet effective approach that fuses the knowledge of source LLMs, which does not require training.Unlike ensemble methods, Cool-Fusion is applicable to any set of source LLMs that have different vocabularies.To overcome the vocabulary discrepancies among LLMs, we ensemble LLMs on text level, allowing them to rerank the generated texts by each other with different granularities.Extensive experiments have been conducted across a variety of benchmark datasets.On GSM8K, Cool-Fusion increases accuracy from three strong source LLMs by a significant margin of 17.4%.

Non-independent and identically distributed (Non-IID) data across edge clients have long posed significant challenges to federated learning (FL) training. Prior works have proposed various methods to mitigate this statistical heterogeneity. While these methods can achieve good theoretical performance, they may lead to the over-correction problem, which degrades model performance and even causes failures in model convergence. In this paper, we provide the first investigation into the hidden over-correction phenomenon brought by the uniform model correction coefficients across clients adopted by the existing methods. To address this problem, we propose TACO, a novel algorithm that addresses the non-IID nature of clients’ data by implementing fine-grained, client-specific gradient correction and model aggregation, steering local models towards a more accurate global optimum. Moreover, we verify that leading FL algorithms generally have better model accuracy in terms of communication rounds rather than wall-clock time, resulting from their extra computation overhead imposed on clients. To enhance the training efficiency, TACO deploys a lightweight model correction and tailored aggregation approach that requires minimum computation overhead and no extra information beyond the synchronized model parameters. To validate TACO’s effectiveness, we present the first FL convergence analysis that reveals the root cause of over-correction. Extensive experiments across various datasets confirm TACO’s superior and stable performance in practice.

The proliferation of Edge Intelligence (EI) and diverse user demands has led to the generation of vast amounts of heterogeneous multimodal data at the network edge. Multimodal Federated Learning (MFL) offers a promising solution for intelligent and personalized services by enabling collaborative training across distributed clients while preserving data privacy. However, existing MFL frameworks remain unsuitable for edge deployment, as they primarily assume homogeneous environments and fail to address heterogeneous client resources. To bridge this gap, we propose$M$ultimodal$F$ederated$E$dge$L$earning (MFEL), a novel paradigm that extends the conventional MFL framework to enable adaptive submodel deployment based on client capabilities. Building on MFEL, we propose MFEL-HAM, a heterogeneous-aware MFL approach that incorporates three core mechanisms: (1) Prototype Networks to align cross-client modality-specific representations, mitigating divergences caused by non-IID data and heterogeneous sensing environments; (2) Rebalanced Modality Gradient Modulation (R-MGM), which adaptively amplifies gradients of underrepresented modalities and suppresses those of dominant ones, alleviating intra-client modality imbalance; and (3) Momentum Knowledge Distillation (MKD), enabling efficient knowledge transfer without sharing raw data, effectively mitigating the impact of resource heterogeneity on collaborative training. Extensive experiments on heterogeneous multimodal datasets show that MFEL-HAM consistently outperforms baselines in accuracy, convergence speed, and training stability, while demonstrating strong generalization across diverse architectures and resource profiles.

The importance of UAV-satellite edge computing networks in disaster relief and scientific exploration has become increasingly prominent, attracting significant attention from both industry and academia. However, under a pre-planned task execution model, fluctuations in data volume often lead to inefficient offloading strategies, significantly increasing the energy consumption risk for UAV-satellite edge computing networks and, in extreme cases, resulting in system failure. Existing offloading approaches either disregard data volume uncertainty, adopt overly conservative robust optimization, or rely on unrealistic distribution assumptions, all of which limit their practicality. To address these limitations, we propose a historical data-driven distributionally robust optimization offloading scheme. Specifically, we first formulate an optimization problem to minimize the total energy consumption and leverage distributionally robust duality theory to derive a tractable formulation. Subsequently, we design an iterative solving algorithm based on the block gradient descent and successive convex approximation methods. Numerical simulations validate that our proposed scheme achieves lower system energy consumption compared to benchmark schemes.

Vertical federated learning (VFL) enables collaborative model training among geo-distributed participants, each with different features of the same samples, but only one party possesses the labels. Communication delays between active and passive parties in VFL significantly hinder its training efficiency. Existing VFL methods adopt asynchronous schemes or multiple local updates per communication round, but they either introduce heavy computation overhead or fail to adapt to dynamic network conditions. This work proposes PASTA, a novel framework employing Adaptive Pipeline Parallelism with Staleness Control for VFL, designed to mitigate these delays and balance training efficiency and model performance. PASTA enables concurrent communication and computation, maximizing resource utilization and minimizing idle time by strategically using stale gradients. Each passive party can send one or more batches of embeddings per communication and conduct stale local training, so that computation times can overlap with communication latency. Since staleness impedes model accuracy despite its benefits in reducing time, a dynamic feedback-based mechanism is proposed to adjust the numbers of embeddings sent and local training iterations based on system heterogeneity. Extensive experiments across various datasets demonstrate that PASTA significantly enhances convergence speed by$1.8 \times$to$4.6 \times$compared to leading VFL systems, without compromising final accuracy. The source code is available at https://github.com/PointerA/PASTA.

With the increasing adoption of unmanned aerial vehicles (UAVs) in critical applications such as infrastructure inspection and emergency response, efficient on-site recognition via live video analytics and streaming has become essential. However, the inherent resource limitation poses significant challenges for performing simultaneous and real-time video analytics and streaming on UAVs. To address this issue, we propose a unified framework that orchestrate the Neural Processing Unit (NPU) and Graph Processing Unit (GPU) of the Systems-on-Chip (SoC) processor to accelerate and carefully schedule the pipeline of video analytics and streaming on UAVs. Additionally, our system incorporates frame interpolation to enable real-time streaming of video analytics results, providing immediate visual feedback to on-site operators. Empirical results on a commercial UAV equipped with Snapdragon 865 SoC platform show that our system reduces per-frame inference latency from 163ms (GPU) to 63ms (NPU), achieving a 2.6× speedup. Combined with optimized pre-processing and frame interpolation, our system increases effective streaming throughput from 2 to 30 FPS, enabling smooth and simultaneous real-time video analytics and streaming.

The growing demand for deploying Artificial Intelligence Generated Content (AIGC) models like Stable Diffusion on resource-constrained edge devices challenges balancing quality, lightweight implementation, and portability. The emergence of WebAssembly (WASM) offers a compactcross-platform, and isolated runtime environment, making it a promising solution for efficient edge AIGC inference. However, current WASM based AI inference solutions are restricted to text interactions, offering limited support for image generation. To solve the challenges, we propose WebAssembly-Rust based WasmSD-Edge, a lightweight, edge-oriented AI image generation framework for high performance on-device Stable Diffusion inference on various edge devices. WasmSD-Edge employs a plugin-based architecture by integrating stable-diffusion.cpp as a WASM backend plugin for the WasmEdge runtime. It exposes a set of WebAssembly System Interfaces (WASI) to support text-to-image, image-to-image, and model convertion. Additionally, a Rust Crate SDK further enables developers to parametrically control inference process and output generation. To evaluate usability and portability of WasmSD-Edge on heterogeneous devices, we deployed it on heterogeneous devices. It achieves high inference speed and image quality with low resource consumption, offering a practical and efficient solution for deploying edge AIGC workflow. The implementation has been merged into WasmEdge — one of the largest WASM community, and source code are available at: https://github.com/WasmEdge/wasmedge-stable-diffusion.

Motivated by the imperative for real-time responsiveness and data privacy preservation, large language models (LLMs) are increasingly deployed on resource-constrained edge devices to enable localized inference. To improve output quality, retrieval-augmented generation (RAG) is an efficient technique that seamlessly integrates local data into LLMs. However, existing edge computing paradigms primarily focus on single-node optimization, neglecting opportunities to holistically exploit distributed data and heterogeneous resources through cross-node collaboration. To bridge this gap, we propose CoEdge-RAG, a hierarchical scheduling framework for retrieval-augmented LLMs in collaborative edge computing. In general, privacy constraints preclude accurate a priori acquisition of heterogeneous data distributions across edge nodes, directly impeding RAG performance optimization. Thus, we first design an online query identification mechanism using proximal policy optimization (PPO), which autonomously infers query semantics and establishes cross-domain knowledge associations in an online manner. Second, we devise a dynamic inter-node scheduling strategy that balances workloads across heterogeneous edge nodes by synergizing historical performance analytics with real-time resource thresholds. Third, we develop an intra-node scheduler based on online convex optimization, adaptively allocating query processing ratios and memory resources to optimize the latency-quality trade-off under fluctuating assigned loads. Comprehensive evaluations across diverse QA benchmarks demonstrate that our proposed method significantly boosts the performance of collaborative retrieval-augmented LLMs, achieving performance gains of 4.23 % to 91.39% over baseline methods across all tasks.

Emerging cyber-physical systems have embraced a large number of IoT devices spanning geo-distributed, which generate and consume massive volumes of data continuously. Accurate and timely spatiotemporal predictions (STP) over these streaming sensor data are critical and, in growing demand, ubiquitous across various edge scenarios such as traffic flow forecasting. Towards that, recent advanced systems have developed sophisticated optimizations among STP pipelines, aiming at optimal prediction performance. However, based on our empirical studies in real-world settings, we identify a previously overlooked bottleneck of end-to-end STP performance: data staleness. To mitigate this issue, in this work, we investigate a new task, namely stream interception, which deliberately terminates the acceptance of incoming sensor data and anticipates model execution with imputed missing features. We propose a novel dynamic interception strategy to determine the time slot to exit waiting and present Grape, an STP system that implements it with practical system designs. Extensive evaluations on real-world traces show that Grape can strike a superior tradeoff between prediction accuracy and serving latency, achieving 1.69-1.90× speedup against traditional all-waiting baselines across various STP services with high prediction accuracy on par with offline optimal cases.

As the adoption of satellite-enabled Internet of Things (IoT) continues to rise, its intricate multi-domain architecture becomes increasingly susceptible to cross-domain cyber threats. Attackers can exploit compromised IoT devices, inject malicious packets into data streams aggregated at the IoT gateway for satellite backhaul, and potentially endanger the satellite network during transmission by exploiting the hardware, software, and protocol vulnerabilities. Compared to single-domain defenses, cooperative defense at the IoT devices, IoT access network, and satellite transmission network provides fine-granularity defense against cross-domain intelligent attacks. However, quantifying cross-domain impacts and tilting incentive misalignment among different participants remain significant challenges, making systematic cooperative defense development a complex task. To address this, we develop a tripartite security game framework to characterize the impacts of attacks and defense methods across both the terrestrial and satellite domains. Leveraging this game model, we devise flow pricing to optimally motivate the IoT Network Operator (IoT-NO) to prevent malicious packet infiltration into the satellite domain. Subsequently, we propose efficient learning algorithms enabling both the IoT-NO to ascertain their ideal flow sampling strategies and the Satellite Service Provider (SAT-SP) to determine optimal flow pricing. The simulation results corroborate the effectiveness of the consolidated game in counteracting cross-domain cyber attacks and facilitating cooperative defense between the IoT-NO and the SAT-SP with non-aligned incentives.

Advances in hardware and AI have enabled edgebased IoT devices to leverage substantial computational and data resources, facilitating large-scale deployment of AI models, particularly through federated learning (FL). However, the high heterogeneity of devices and resource contention at the edge make collaborative optimization of resource scheduling for multiple FL tasks challenging. To tackle this, we propose a novel Multi-Task Asynchronous Federated Learning (MTAFL) architecture, which enhances resource utilization efficiency by enabling orthogonal multiplexing of computation and communication resources through adjusting local epochs on edge devices. Then, we formulate an optimization problem in the MTAFL framework to manage resources and local epochs, aiming to minimize energy consumption while achieving FL performance. However, intricate couplings between resource allocation and local control complicate the long-term FL process. To address this, we employ a two-step relaxation approach and develop an efficient optimization strategy based on the block coordinate descent algorithm. To enhance optimization granularity, we extend the MTAFL framework by incorporating device-level data characteristics. We propose a Gaussian Process-based client selection mechanism that dynamically characterizes and predicts training loss trajectories across clients. After selecting clients for each task, we optimize resource allocation and local control strategies in the system. Extensive numerical evaluations corroborate the superior performance of the proposed approaches over existing schemes.

With the emergence of mobile edge applications and the demand for access-as-a-service, satellite mobile edge computing stands out as a disruptive technology for delivering low-latency edge service. In this article, we focus on service deployment to the edge satellites for terrestrial users, which is a key enabling technology in satellite mobile edge computing and will replace traditional centralized cloud computing. Most existing works on service deployment consider a centralized nonconvex optimization problem with high computational overhead. However, in practice, it is difficult for a single satellite to solve computationally expensive network optimization problems. To this end, we propose a distributed optimization model based on the alternating direction method of multipliers (ADMMs), which can relieve the computational burden by leveraging collaborative calculations among multiple satellites. Our proposed model minimizes the total delay of service deployment for terrestrial users by formulating a joint optimization problem that involves deployment decisions, CPU resource decisions, transmission decisions, and caching decisions. Furthermore, we propose a novel approximation method that transforms the nonconvex optimization problem to a convex one to make the joint optimization problem solvable in polynomial time. Finally, we conduct experiments using scaled global population data and show that the proposed distributed model outperforms the baselines.

With the rapid development of on-board computing technology, on-orbit information processing has become a new direction for reducing service response delays and improving the quality of space-based information services. Especially in space-air–ground integrated applications in 6G networks, remote sensing image processing tasks are highly important because of their critical role in applications such as environmental monitoring and public safety. However, the fluctuations in data volume due to significant scene differences, along with the limitations in individual satellite capabilities caused by size and power constraints, present new challenges for on-orbit image processing. To address these challenges, we model a data-driven on-orbit resource scheduling problem for space-air-ground integrated networks based on distributionally robust optimization, aiming to minimize the average image processing delay. We first construct an ambiguity set based on the Wasserstein distance and the historical distribution of image data, which helps transform the original upper-bound expectation problem into an explicitly expressed mixed-integer nonlinear (MINLP) problem. Furthermore, to reduce complexity and expedite the solution process, we decouple the MINLP problem into three subproblems using the block coordinate descent method and designed an iterative solving algorithm. The numerical results demonstrate that our proposed method achieves better fitting accuracy than traditional methods and reduces the average image processing delay.

Federated learning (FL) has been extensively applied in industrial cyber-physical systems (ICPSs) to develop powerful models for complex industrial tasks (e.g., fault diagnosis), while safeguarding industrial data confidentiality. Asynchronous federated learning (AFL) effectively mitigates the straggler issue in the synchronous paradigm by aggregating client models in a first-come-first-served manner. Proper client selection is crucial for achieving efficient model training in AFL. A widely adopted model for implementing client selection in AFL is multi-armed bandit (MAB), which models client selection as arm pulling. Existing MAB-based client selection schemes overlook practical scenarios where direct client-server communications are unfavorable or unavailable (for example, in ICPSs such as mines, where communication infrastructure is underdeveloped, direct client-server communication is often unreliable or even unfeasible). In such cases, the server needs to incentivize self-interested relays to perform arm-pulling actions, including selecting the right client and relaying the communication from the selected client to the server. This paper proposes the first framework of incentivized online client selection for AFL. The design and optimization of such a framework involve significant challenges due to the tight coupling between unknown client behavior and private relay cost. To circumvent this challenge, we adopt the dual-based method and construct a special Lagrangian function that incorporates client behavior learning and relay cost revelation, and utilize it to design a socially-optimal mechanism for the framework. Our mechanism satisfies several desirable properties, including voluntary participation, incentive compatibility, relay utilization fairness, and client participation fairness. The proposed mechanism achieves the same asymptotic performance as the state-of-the-art benchmark that requires additional information. Furthermore, our analysis reveals that more available relays bring our mechanism closer to the theoretical upper bound of social performance. Numerical results demonstrate that our proposed mechanism achieves up to 85% and 99% of the social welfare obtained by the benchmarks.

Mobile Edge Computing (MEC) revolutionizes real-time applications by extending cloud capabilities to network edges, enabling efficient computation offloading from mobile devices. In recent years, the location privacy concern within MEC offloading has been recognized, prompting the proposal of various methodologies to mitigate this concern. However, this paper demonstrates that the prevailing privacy protection methods exhibit vulnerabilities. First, we analyze the shortcomings of current methodologies through both system modeling and evaluation metrics. Then, we introduce a Learning-based Trajectory Reconstruction Attack (LTRA) to expose the weaknesses, achieving up to 91.2% reconstruction accuracy against the state-of-the-art protection method. Further, based onw-event differential privacy, we propose an ℓ-trajectory differentially private mechanism, i.e., OffloadingBD. Compared to the existing works, OffloadingBD provides more flexible and enhanced protection with sound privacy theoretical guarantee. Lastly, we conduct extensive experiments to evaluate LTRA and OffloadingBD. The experiment results show that LTRA has good generalization ability and OffloadingBD showcases a superior balance between privacy and utility compared with baselines.

In the forthcoming era of 6G networks, delay-sensitive applications for Internet of Things (IoT) are poised to become the prevailing services with ultra-reliable and low-latency (URLLC) requirements. Unlike traditional video caching, IoT-based edge caching faces unique challenges due to diverse data types, update frequencies, and computational needs, requiring integrated storage and computational resource management. To support the more stringent requirements for these innovative applications, mobile edge computing (MEC) is introduced to enhance the service reliability of delay-sensitive applications in the 6G era. However, task offloading, as an indispensable procedure in MEC, would encounter many challenges, such as network jitter and resource insufficiency, possibly leading to unpredictable queuing delays and other negative issues. To ensure reliable services in a dynamical MEC environment, the caching-enabled MEC network has emerged as a novel architecture, placing computing and storage resources in the edge network. In this paper, we investigate the caching-enabled MEC to support reliable task offloading for delay-sensitive applications, with a focus on IoT scenarios. In our system model, we formulate the task process as a two-hop tandem queuing system with limited capacity, including task transmission and computation queues. The Martingale theory is leveraged to analyze the delay violation probability in this system, demonstrating how the offloading and caching decisions affect the end-to-end (E2E) delay. Besides, task offloading and resource allocation policies are integrated to reduce high system costs, including energy consumption and cache resource rental costs. Based on the delay analysis of martingale theory, we propose an advanced deep reinforcement learning (DRL) algorithm called Dynamic Request Aware Soft Actor-Critic (DRA-SAC) algorithm to achieve minimal system costs by obtaining the optimal task offloading and resource allocation policies, including caching and computation resources. We conduct some illustrative studies to evaluate the proposed scheme. The algorithm we have put forward outperforms benchmark algorithms regarding both cache hit ratio and system cost.

Privacy-preserving vector mean estimation is a crucial primitive in federated analytics. Existing practices usually resort to Local Differentiated Privacy (LDP) mechanisms that inject random noise into users’ vectors when communicating with users and the central server. Due to the privacy-utility trade-off, the privacy budget has been widely recognized as the bottleneck resource that requires well-provisioning. In this paper, we explore the possibility of privacy budget recycling and propose a novelChainDPframework enabling users to carry out data aggregation sequentially to recycle the privacy budget. We establish a sequential game to model the user interactions in our framework. We theoretically show the mathematical nature of the sequential game, solve its Nash Equilibrium, and design an incentive mechanism with provable economic properties. To alleviate potential privacy collusion attacks, we further derive a differentially privacy-guaranteed protocol to avoid holistic exposure. Our numerical simulation validates the effectiveness of ChainDP, showing that it can significantly save privacy budget as well as lower estimation error compared to the traditional LDP mechanism.

Benefiting from hardware upgrades and deep learning techniques, more and more end devices can independently support a variety of intelligent applications. Further powered by edge computing technologies, the end-edge collaboration paradigm becomes one mainstream approach for achieving advanced edge intelligence (EI). To fully exploit the system resources, it is desirable to coordinate diverse EI services efficiently. Thus, we present a novel framework to jointly optimize the cost-performance trade-off for two distinct but typical EI services, where end devices simultaneously perform federated learning (FL) model training and conduct model inference with the assistance of edge offloading. However, balancing the long-term cost-performance trade-off is highly non-trivial, especially in the absence of knowledge of future system dynamics. Moreover, the capacity heterogeneity further increases the difficulty of service coordination among resource-limited end devices. To overcome these challenges, we first analyze the optimality of inference offloading decisions with and without FL model training and quantify their mutual effects due to local resource contention. By incorporating the loss estimation of FL training model, we then propose a novel proactive policy with theoretical guarantees, which proactively controls the stopping of FL training procedure to balance well the trade-offs between FL model performance and resource costs while fulfilling the inference performance requirements. Extensive results show the efficiency and robustness of our proposed algorithm for EI service coordination in dynamic end-edge collaboration scenarios.

With the rapid proliferation of mobile devices, a growing number of intelligent applications are being deployed at the network edge, placing immense strain on the processing capabilities of edge computing. Therefore, resourceconstrained edge servers frequently experience overload due to highly dynamic workloads. To address this, one approach involves forwarding user requests to the cloud or other edge servers, albeit at the cost of increased transmission latency. Alternatively, dynamic scaling of edge clusters can be employed to enhance processing capacity, thereby mitigating latency but at the expense of additional service configuration and hosting expenses. By integrating their complementary benefits, we study the joint optimization problem of dynamic scaling and request routing within a multi-edge cluster collaborative framework, which fully exploits cluster resources to manage the temporal and spatial varying edge workloads. This collaborative framework aims to minimize overall request latency while satisfying an acceptable time-averaged budget cost. However, the complex coupling between scaling and routing decisions, along with the uncertainty of future system information (e.g., user request workloads) impedes the derivation of an optimal offline policy over the long term. Thus, considering the different decision granularities, we employ the two-timescale Lyapunov optimization technique to decouple the original problem into a series of independent online optimization problems with the current system state. In particular, we make cluster scaling decisions in each large timescale and request routing decisions in each small timescale. Given that the decoupled large-timescale subproblems involve NP-hard mixed-integer linear programming, we design an edge resource-aware greedy rounding algorithm to efficiently produce approximate optimal solutions. Finally, both rigorous theoretical analysis and extensive trace-driven evaluations demonstrate the superiority of our proposed algorithm over its counterparts.

Large Language Models (LLMs) have demonstrated remarkable capabilities, leading to a significant increase in user demand for LLM services. However, cloud-based LLM services often suffer from high latency, unstable responsiveness, and privacy concerns. Therefore, multiple LLMs are usually deployed at the network edge to boost real-time responsiveness and protect data privacy, particularly for many emerging smart mobile and IoT applications. Given the varying response quality and latency of LLM services, a critical issue is how to route user requests from mobile and IoT devices to an appropriate LLM service (i.e., edge LLM expert) to ensure acceptable quality-of-service (QoS). Existing routing algorithms fail to simultaneously address the heterogeneity of LLM services, the interference among requests, and the dynamic workloads necessary for maintaining long-term stable QoS. To meet these challenges, in this paper we propose a novel deep reinforcement learning (DRL)-based QoS-aware LLM routing framework for sustained high-quality LLM services. Due to the dynamic nature of the global state, we propose a dynamic state abstraction technique to compactly represent global state features with a heterogeneous graph attention network (HAN). Additionally, we introduce an action impact estimator and a tailored reward function to guide the DRL agent in maximizing QoS and preventing latency violations. Extensive experiments on both Poisson and real-world workloads demonstrate that our proposed algorithm significantly improves average QoS and computing resource efficiency compared to existing baselines.

Transformer-based models have unlocked a plethora of powerful intelligent applications at the edge, such as voice assistant in smart home. Traditional deployment approaches offload the inference workloads to the remote cloud server, which would induce substantial pressure on the backbone network as well as raise users' privacy concerns. To address that, in-situ inference has been recently recognized for edge intelligence, but it still confronts significant challenges stemming from the conflict between intensive workloads and limited on-device computing resources. In this paper, we leverage our observation that many edge environments usually comprise a rich set of accompanying trusted edge devices with idle resources and proposeGalaxy+, a collaborative edge AI system that breaks the resource walls across heterogeneous edge devices for efficient Transformer inference acceleration.Galaxy+introduces a novel hybrid model parallelism to orchestrate collaborative inference, along with a heterogeneity and memory-aware parallelism planning for fully exploiting the resource potential. To mitigate the impact of tensor synchronizations on inference latency under bandwidth-constrained edge environments,Galaxy+devises a tile-based fine-grained overlapping of communication and computation. Furthermore, a fault-tolerant re-scheduling mechanism is developed to address device-level resource dynamics, ensuring stable and low-latency inference. Extensive evaluation based on prototype implementation demonstrates thatGalaxy+remarkably outperforms state-of-the-art approaches under various edge environment setups, achieving a$1.2\times$to$4.24\times$end-to-end latency reduction. Besides,Galaxy+can adapt to device-level resource dynamics, swiftly rescheduling and restoring inference in the presence of unexpected straggler devices.

Edge-cloud computing networks integrate dispersed computing resources of edges and clouds through networks, which improves resource utilization by flexibly scheduling tasks to suitable computing nodes. The performance of edge-cloud computing networks depends significantly on the amount of computing resources and the task scheduling scheme. In this work, we propose a novel computing resource trading and task scheduling framework for edge-cloud computing networks with arbitrary network topology. Specifically, we consider a third-party platform which incentivizes computing nodes to share computing resources by designing proper resource pricing mechanisms, and charges customers execution fees by scheduling tasks optimally in the edge-cloud computing network. The platform’s resource pricing and task scheduling optimization problem captures the unique features of edge-cloud computing networks including the heterogeneities of computing resources and tasks, as well as the multi-hop offloading in arbitrary topology, which is challenging to solve. We solve the problem for the homogeneous workload scenario and the heterogeneous workload scenario, respectively. For the homogeneous workload scenario, we propose a multi-round proposer-voter algorithm (MPV) that achieves the global optimum in polynomial time for the non-competitive case. For the heterogeneous workload scenario, we first propose a Gibbs sampling based iterative algorithm (GSI), which updates task scheduling strategies iteratively using Gibbs sampling and converges to the global optimum with high probability. We further propose a distributed alternating update algorithm (DAU), which converges to the local optimum in a distributed manner with linear complexity. Numerical results demonstrate the effectiveness of our proposed resource trading and task scheduling schemes.

Graph-structured data applications promote the development of Graph Neural Networks (GNN) in recent years. Due to privacy concerns, collecting graph data stored in massive client devices for centralized graph learning is prohibitive. It is natural to integrate federated learning (FL) in graph learning to address this issue, which enables clients to collaborate on training a shared model without uploading their data. This generates an emerging paradigm of federated graph learning (FGL). However, due to various costs incurred by FGL training, the collaboration between the server and clients is still a challenging issue in FGL, which remains largely unexplored in existing studies. To bridge this gap, we propose a cost-efficient collaboration framework for FGL of graph convolutional networks on semi-supervised node classification tasks, i.e., Joint Client and Cross-Client Edge Selection (JC3ES) for the server. Specifically, we first characterize how varies graph structure affect the final convergence performance of the FGL model. We then reveal the fundamental supermodular property in client selection. Based on this, we further devise an approximately optimal algorithm for the server and theoretically derive the performance gap between the proposed algorithm and the optimal solution. Extensive numerical evaluations show that our proposed algorithm achieves outstanding performance in cost-efficient collaboration for FGL on popular graph datasets.

Due to the penetration of edge computing, a wide variety of workloads are sunk down to the network edge to alleviate huge pressure of the cloud. With the presence of high input workload dynamics and intensive edge resource contention, it is highly non-trivial for an edge proxy to optimize the scheduling of heterogeneous services with diverse QoS requirements. In general, online services should be quickly completed in a quite stable running environment to meet their tight latency constraint, while offline services can be processed loosely for their elastic soft deadlines. To well coordinate such services at the resource-limited edge cluster, in this paper, we study an edge-centric resource provisioning optimization for dynamic online and offline services co-location, where the proxy seeks to maximize timely online service performances while maintaining satisfactory long-term offline service performances. However, intricate hybrid couplings for provisioning decisions arise due to heterogeneous constraints of the co-located services and their different time-scale performances. We hence first propose a reactive provisioning approach without requiring a prior knowledge of future system dynamics, which leverages a Lagrange relaxation for devising constraint-aware stochastic subgradient algorithm to deal with the challenge of hybrid couplings. To further boost the performance by integrating powerful machine learning techniques, we then advocate a predictive provisioning approach, where future request arrivals can be estimated accurately. To align with practical deployments, we incorporate a tunable prediction window mechanism, which well balances the potential improvement and degradation of online performance in imperfect prediction scenarios. With rigorous theoretical analysis and extensive trace-driven evaluations, we show the superior performance of our proposed algorithms for online and offline services co-location at the edge.

Transformer models have shown significant success in a wide range of tasks. However, the massive resources required for its inference prevent deployment on a single device with relatively constrainted resources, thus leaving a high threshold of integrating their advancements. Observing scenarios such as smart home applications on edge devices and cloud deployment on commodity hardware, it is promising to distribute Transformer inference across multiple devices. Unfortunately, due to the tightly-coupled feature of Transformer model, existing model parallelism approaches necessitate frequent communication to resolve data dependencies, making them unacceptable for distributed inference, especially under relatively weak interconnection. In this paper, we propose DeTransformer, a communication-efficient distributed Transformer inference system. The key idea of DeTransformer involves the co-design of Transformer architecture to reduce the communication during distributed inference. In detail, DeTransformer is based on a novel block parallelism approach, which restructures the original Transformer layer with a single block to the decoupled layer with multiple sub-blocks. Thus, it can exploit model parallelism between sub-blocks. Next, DeTransformer contains an adaptive execution approach that strikes a trade-off among communication capability, computing power and memory budget over multiple devices. It incorporates a two-phase planning for execution, namely static planning and runtime planning. The static planning runs offline, containing a profiling procedure and a weight placement strategy before execution. The runtime planning dynamically determines the optimal parallel computing strategy from an expertly crafted search space based on real-time requests. Notably, this execution approach can adapt to heterogeneous devices by distributing workload based on devices’ computing capabilities. We conduct experiments for both auto-regressive and auto-encoder tasks of Transformer models. Experimental results show that DeTransformer can reduce distributed inference latency by up to 2.81× compared to the SOTA approach on 4 devices, while effectively maintaining task accuracy and a consistent model size.

The increasing demand for connectivity in 5G networks has led to a focus on massive machine-type communication (mMTC) in mobile edge computing (MEC) for IoTs. However, the proliferation of IoT devices has resulted in densely deployed networks and led to a high volume of task offloading to the same edge servers simultaneously. As a consequence, mMTC applications may experience service congestion, negatively impacting service reliability. To enhance the service reliability of latency-sensitive applications, task replication with resource allocation is proposed in MEC, in which a task can be sent simultaneously to multiple computing nodes. Task replication can reduce task latency and improve service reliability at the cost of consuming more computation resources. However, unconstrained task replication may result in too many uploading links, leading to severe costs in network operation. To handle the above challenge, we propose a constrained stochastic optimization problem by task replication with wireless resource block (RB) allocation and edge server queue management. To ensure queue stability while minimizing cost, we design one strategy based on the Lyapunov optimization framework. Accordingly, we further model RB allocation as a mean-field game (MFG) due to the intensive coupling of the RB pool for massive users. Tractable partial differential equations are used to analyze MFG equilibrium, and we derive the optimal edge server queue management based on a given task replication strategy and RB allocation scheme. Our theoretical analysis demonstrates that our algorithm closely approaches the optimal overall costs within a small gap, and simulation results show that our strategy generates a significantly lower cumulative cost than other alternative strategies.

In cross-silo federated learning (FL), clients of common interest cooperatively train a global model without sharing local sensitive data, but they still face potential privacy leakage due to privacy threats from malicious attackers. Although some articles have proposed effective privacy-preserving mechanisms for FL (such as differential privacy (DP)), clients in cross-silo FL are usually different companies or organizations who may behave selfishly to optimize their own benefits. In this article, we study DP-based cross-silo FL where clients selfishly decide their participation levels (i.e., data sizes for model trainings) and privacy leakage tolerance levels to trade off between model accuracy loss and privacy loss, and we model clients’ interactions as a participation-dependent privacy preservation game. It is challenging to analyze the game since the comprehensive impact of participation levels and privacy leakage tolerance levels on model accuracy is unclear and the behaviors of heterogeneous clients are coupled in a highly complex manner. To capture the impact of participation and privacy preservation behaviors, we first characterize the optimality gap of DP-based cross-silo FL for both convex and non-convex models, where the privacy leakage tolerance levels and the participation levels are coupled nonlinearly. We model clients’ costs based on the optimality gap, and prove that clients’ selfish participation-dependent privacy preservation game is a potential game. To analyze the optimal strategies of heterogeneous clients in a stable state, we derive the closed-form expression for the unique Nash equilibrium (NE), where clients may choose full participation or partial participation, and the equilibrium privacy preservation strategy depends on clients’ accuracy-privacy preference ratios. We analyze the social efficiency of the NE by calculating the price of anarchy (PoA) and show that the PoA increases with the number of clients and the heterogeneity of clients’ model accuracy preferences. To improve the social efficiency achieved at equilibrium, we design a socially efficient incentive mechanism that allows clients with large model accuracy preferences to compensate clients with small model accuracy preferences. Extensive experiments verify our theoretical results for both the convex and non-convex models as well as both the i.i.d. data distribution case and the non-i.i.d. data distribution case.

Transformer models have shown significant success in a wide range of tasks. Meanwhile, massive resources required by its inference prevent scenarios with resource-constrained devices from in-situ deployment, leaving a high threshold of integrating its advances. Observing that these scenarios, e.g. smart home of edge computing, are usually comprise a rich set of trusted devices with untapped resources, it is promising to distribute Transformer inference onto multiple devices. However, due to the tightly-coupled feature of Transformer model, existing model parallelism approaches necessitate frequent communication to resolve data dependencies, making them unacceptable for distributed inference, especially under weak interconnect of edge scenarios. In this paper, we propose DeTransformer, a communication-efficient distributed in-situ Transformer inference system for edge scenarios. DeTransformer is based on a novel block parallelism approach, with the key idea of restructuring the original Trans-former layer with a single block to the decoupled layer with multi-ple sub-blocks and exploit model parallelism between sub-blocks. Next, DeTransformer contains an adaptive placement approach to automatically select the optimal placement strategy by striking a trade-off among communication capability, computing power and memory budget. Experimental results show that DeTransformer can reduce distributed inference latency by up to 2.81 x compared to the SOTA approach on 4 devices, while effectively maintaining task accuracy and a consistent model size.

Federated Distillation (FD), a popular variant of Federated Learning (FL), has attracted researchers' attention due to its ability to support heterogeneous model training. Generally, FD allows clients to upload logits associated with public datasets for knowledge transfer, yet logits may pose privacy risks. In this study, we provide the first demonstration of the impact of privacy risks caused by logits. Specifically, we design a data reconstruction attack against logits named L-Attack which can reveal sensitive information about the target client without access to the target model. Via the zeroth-order optimization technique, L-Attack involves training a server-side generator that unveils certain features of private data owned by the target client. To defend against L-Attack, we propose a label aggregation-based FD algorithm called LabelAvg which allows clients to upload predicted hard labels for knowledge transfer instead of logits. Due to the insufficient information in labels for distillation, LabelAvg provides a voting-based label smoothing mechanism that enables the server to construct smooth labels from received labels. The generated smooth labels which stand for the consensus among all clients, indicate the approximate probability distribution. Thus, these smoothed labels bear a striking similarity to logits and can be used for distillation. Analysis and experimental results prove LabelAvg is superior to baselines in terms of accuracy, privacy, and communication data volume.

Federated Learning (FL) is a distributed machine learning paradigm that achieves a globally robust model through decentralized computation and periodic model synthesis, primarily focusing on the global model’s accuracy over aggregated datasets of all participating clients. Personalized Federated Learning (PFL) instead tailors exclusive models for each client, aiming to enhance the accuracy of clients’ individual models on specific local data distributions. Despite of their wide adoption, existing FL and PFL works have yet to comprehensively address the class-imbalance issue, one of the most critical challenges within the realm of data heterogeneity in PFL and FL research. In this paper, we propose FedReMa, an efficient PFL algorithm that can tackle class-imbalance by 1) utilizing an adaptive inter-client co-learning approach to identify and harness different clients’ expertise on different data classes throughout various phases of the training process, and 2) employing distinct aggregation methods for clients’ feature extractors and classifiers, with the choices informed by the different roles and implications of these model components. Specifically, driven by our experimental findings on inter-client similarity dynamics, we develop critical co-learning period (CCP), wherein we introduce a module named maximum difference segmentation (MDS) to assess and manage task relevance by analyzing the similarities between clients’ logits of their classifiers. Outside the CCP, we employ an additional scheme for model aggregation that utilizes historical records of each client’s most relevant peers to further enhance the personalization stability. We demonstrate the superiority of our FedReMa in extensive experiments. The code is available at https://github.com/liangh68/FedReMa.

Service Function Chaining (SFC) is a fundamental technology for resource management in Space-Air-Ground Integrated Network (SAGIN). The heterogeneity and dynamic nature of network resources in SAGIN increase the complexity of SFC-based resource allocation. However, existing work rarely considers the issue of economically efficient resource allocation under cost constraints. To solve the issue, this study explicitly analyzes resource characteristics and establishes a cost-driven online auction mechanism for SFC allocation. First, we formulate a novel SFC allocation problem for SAGIN, aiming at maximizing social welfare while considering operational costs. We then adopt Fenchel duality to convert the primal problem into a dual problem and design a payment strategy that facilitates the dynamic updating of resource marginal prices. Our algorithm achieves optimal SFC allocation and pricing outcomes while guaranteeing bidding truthfulness, individual rationality, and polynomial-time complexity. Finally, we validate the online auction’s competitiveness through rigorous theoretical analysis and simulation studies driven by real-world traces.

Federated learning is a popular distributed machine learning paradigm that keeps data locally at clients. To further enhance privacy protection, differential privacy techniques are incorporated in the federated learning framework. We can quantify the privacy budget (or privacy protection level) through differential privacy and allocate the budget to different communication rounds according to the composition property of differential privacy. Recent works have shown that suitably allocating budgets to different iterations can improve model performance. How to allocate privacy budgets in different communication rounds for different clients in the federated learning framework is a significant problem to study. The problem is challenging to solve due to the unknown relationship between noise levels and the model accuracy and the coupling property of the clients' decisions. In this paper, we propose a method based on multi-agent reinforcement learning to solve the privacy budget allocation problem, which maximizes the accuracy of the federated learning model given limited privacy budgets for the clients. The experiments show that our proposed method is better than the uniform allocation, arithmetic sequence allocation, and exponential allocation methods.

Federated learning (FL), a decentralized machine learning approach, enables privacy-preserving and collaborative model training without centralizing sensitive data. It has been successfully applied in various domains, including e-commerce, healthcare, and finance. However, existing FL schemes often fail to address personalized task requirements, such as prior-itizing the accuracy of specific classes within a dataset. The recent surge in Artificial Intelligence Generated Content (AIGC) offers potential to meet these personalized requirements by augmenting the training data of specific classes with generative models. Nevertheless, integrating generative models with FL introduces challenges, such as non-compliant data, disorganized distributions, and limited computing power on edge devices. To address these challenges, we propose AIGC-augmented Federated Preference Learning (FPL), which focuses on training specific data classes, referred to as preference classes (PCs). To improve the quality of AI -generated data, we implement strategies such as pre-training and fine-tuning across various datasets. Additionally, we enhance FL efficiency through a client selection strategy that matches generated data tasks with suitable clients and an AIGC data distribution strategy that optimally allocates data where it is most needed. We validate the feasibility and effectiveness of AIGC-augmented FPL by conducting experiments on the MNIST and CIFAR-10 datasets from various perspectives.

Video analytics is considered the killer application of edge computing and has been successfully deployed across diverse domains. Yet, executing video analytics on mobile devices presents notable challenges owing to the considerable computational demands and frame rate requirements of DNN models. Current mobile inference frameworks often concentrate on enhancing model inference performance on the CPU or GPU, overlooking the potential of the Digital Signal Processor (DSP) – an emerging heterogeneous processor increasingly integrated into modern mobile processors. In this paper, we introduce COUPLE, an orchestration framework for video analytics on heterogeneous mobile processors, with the goal of optimizing real-time video analysis through the collaboration of CPU, GPU and DSP. To tackle the accuracy loss of DSP inference, we introduce the Anchor Frame Calibration mechanism, utilizing high-precision GPU inference results and frame similarities to mitigate accuracy loss on the DSP. Additionally, we design a lightweight progressive scheduler to distribute video frames to GPU and DSP, maximizing inference Average Precision (AP) under performance (i.e., frame rate) and power constraints. COUPLE has been implemented on the Qualcomm's Snapdragon 888 mobile SoC, extensive evaluation results demonstrate its efficacy in imnroving the inference performance and accuracy.

Large language models (LLMs) have unlocked a plethora of powerful applications at the network edge, such as intelligent personal assistants. Data privacy and security concerns have prompted a shift towards edge-based fine-tuning of personal LLMs, away from cloud reliance. However, this raises issues of computational intensity and resource scarcity, hindering training efficiency and feasibility. While current studies investigate parameter-efficient fine-tuning (PEFT) techniques to mitigate resource constraints, our analysis indicates that these techniques are not sufficiently resource-efficient for edge devices. Other studies focus on exploiting the potential of edge devices through resource management optimization, yet are ultimately bottlenecked by the resource wall of individual devices.

Transformer-based models have unlocked a plethora of powerful intelligent applications at the edge, such as voice assistant in smart home. Traditional deployment approaches offload the inference workloads to the remote cloud server, which would induce substantial pressure on the backbone network as well as raise users’ privacy concerns. To address that, in-situ inference has been recently recognized for edge intelligence, but it still confronts significant challenges stemming from the conflict between intensive workloads and limited on-device computing resources. In this paper, we leverage our observation that many edge environments usually comprise a rich set of accompanying trusted edge devices with idle resources and propose Galaxy, a collaborative edge AI system that breaks the resource walls across heterogeneous edge devices for efficient Transformer inference acceleration. Galaxy introduces a novel hybrid model parallelism to orchestrate collaborative inference, along with a heterogeneity-aware parallelism planning for fully exploiting the resource potential. Furthermore, Galaxy devises a tile-based fine-grained overlapping of communication and computation to mitigate the impact of tensor synchronizations on inference latency under bandwidth-constrained edge environments. Extensive evaluation based on prototype implementation demonstrates that Galaxy remarkably outperforms state-of-the-art approaches under various edge environment setups, achieving up to 2.5× end-to-end latency reduction.

Rapid advances in low Earth orbit (LEO) satellite technology and satellite edge computing (SEC) have facilitated a key role for LEO satellites in enhanced Earth observation missions (EOM). These missions (e.g., remote object detection) typically require multi-satellite cooperative observations of a large region of interest (RoI) area, as well as the observation image routing and computation processing, enabling accurate and real-time responsiveness. However, optimizing the resources of LEO satellite networks is nontrivial in the presence of its dynamic and heterogeneous properties. To this end, we propose SECO, a SEC-enabled framework that jointly optimizes multi-satellite observation scheduling, routing and computation node selection for enhanced EOM. Specifically, in the observation phase, we leverage the orbital motion and the rotatable onboard cameras of satellites, and propose a distributed game-based scheduling strategy to minimize the overall size of captured images while ensuring full (observation) coverage. In the sequent routing and computation phase, we first adopt image splitting technology to achieve parallel transmission and computation. Then, we propose an efficient iterative algorithm to jointly optimize image splitting, routing and computation node selection for each captured image. On this basis, we propose a theoretically guaranteed systemwide greedy-based strategy to reduce the total time cost (i.e., transmission, computation and queuing delay) over simultaneous processing for multiple images. Extensive experiments based on real-world datasets demonstrate that SECO can achieve up to a 60.7% reduction in overall time cost compared to baselines.

Today’s video streaming service providers have exploited cloud-edge collaborative networks for geo-distributed video delivery. The existing content delivery network (CDN) scheduling and adaptive bitrate algorithms may not fully utilize edge resources or lack a global control to optimize resource sharing. The emerging super-resolution (SR) approach can unleash the potential of leveraging computation resources to compensate for bandwidth consumption, by producing high-quality videos from low-resolution contents. Yet the uncertain SR resource sensitivity and its interplay with bitrate adaptation are underexplored. In this work, we propose Rosevin, the first resource scheduler that jointly decides the bitrates and fine-grained resource allocation to perform SR at the edge, which can learn to optimize the long-term QoE for distributed end users. To handle the time-varying and complex space of decisions as well as a non-smooth objective function, Rosevin realizes a novel online combinatorial learning algorithm, which nicely integrates convex optimization theories and online learning techniques. In addition to theoretically analyzing its performance, we implement an SR-assisted video streaming prototype of Rosevin and demonstrate its advantages over several video delivery benchmarks.

Driven by the continuous upgrading of hardware devices, a notable shift from traditional single-FL task to complicated multi-FL tasks is emerging in edge computing, supporting richer intelligent services. With the presence of high capacity heterogeneity and intensive resource contention at the edge, it is highly non-trivial to collaboratively optimize resource scheduling of multiple FL tasks with diverse QoS requirements. To well tackle the above challenge, we firstly propose a novel Multi-Task Asynchronous Federated Learning (MTAFL) architecture. This novel framework has the potential to enhance resource utilization efficiency by enabling the orthogonal multiplexing of computation and communication resources through adjusting the number of local epochs on edge clients. Then, we formulate an optimization problem within the MTAFL framework, aiming at managing resource allocation and client scheduling to minimize the system-wide energy consumption while achieving the target FL performance. However, intricate couplings for resource allocation and local training decisions arise during the long-term FL process. We hence employ a two-step relaxation approach to transform original non-convex problem into a multi-convex problem, and further devise an efficient optimization strategy based on the block coordinate descent algorithm. Extensive numerical evaluations corroborate the superior performance of the proposed MTAFL framework over existing schemes.

Federated Learning (FL) has surged in popularity, allowing machine learning models to be collaboratively trained using decentralized client data, all while upholding privacy and security standards. However, leveraging locally-stored data introduces challenges related to data heterogeneity. While many past studies have addressed this non-IID problem, they often overlook the dynamic nature of each individual client’s data or disrupt its continuous shift. In this paper, our emphasis is on the challenges posed by temporal data distribution shift alongside non-IID data across clients, a more prevalent yet complex situation in real-world FL. We propose to analytically capture the evolving nature of each local data distribution, by modeling them as a time-varying composite of multiple latent Gaussian distributions. We then employ the expectation maximization (EM) algorithm to deduce the distribution model parameters based on the prevailing observed training data, ensuring that the learned mixture proportion weights mirror a consistent trajectory. Additionally, by embedding an adaptive data partitioning method into the EM algorithm and using each partition to train a distinct sub-model, we realize an intuitive and novel personalized FL paradigm. This refines the FL training by exploiting the heterogeneity and temporal shifts of clients’ datasets. We derive analytical results to guarantee the convergence of our training method. Comprehensive tests across diverse datasets and distribution configurations also underscore our enhanced efficacy compared to several state-of-the-art.

The deep integration of federated learning (FL) and edge computing holds great promise in delivering ubiquitous edge AI services. However, in light of the upcoming carbon peaking and neutrality era, existing research has largely overlooked the sustainability challenges of FL in future edge computing. Therefore, we first propose a novel carbon-efficient FL framework in this paper, which leverages client sampling and model pruning approaches to adjust carbon-aware local model training during the long-term FL procedure, adapt to heterogeneous and dynamic edge environments, such as time-varying renewable energy and edge workloads. We then conduct a theoretical analysis of its convergence bound, based on which we introduce an online control algorithm to efficiently balance the trade-off between training performance and carbon emission, i.e., maximizing carbon efficiency while ensuring satisfactory FL performance. The effectiveness of proposed algorithm is verified by extensive trace-driven simulations, reducing up-to 72% carbon emissions than other methods.

Given the high mobility and sensor-carrying capability, vehicle crowdsensing (VCS) has become a significant part of urban crowdsensing tasks in the development of smart cities. Ride-hailing vehicles, which are widely distributed in cities, can be a powerful tool for carrying out VCS. However, dispatching the vehicles to jointly benefit VCS and order serving is challenging, as the goals of these two tasks may not be consistent or even conflict. The distribution of ride orders and the distribution of point-of-interests (PoIs) may not coincide in time and geography. In addition, these orders and data PoIs have distinct forms of timeliness: prolonged waiting makes orders invalid and data with a larger age-of-information (AoI) has lower utility. We propose an online framework by extending multi-agent reinforcement learning (MARL) with careful augmentation to optimize the profit of order-serving and the data utility of crowdsensing. A new quality-of-service (QoS) metric is designed to characterize the utility of the two joint tasks, and formal mathematical modeling drives our MARL design. In particular, we integrated graph neural networks (GNN) to enhance state representations and capture the graph-structured dependencies among vehicles. We developed a simulator and conducted extensive experiments utilizing the New York City Taxi dataset. Experimental results demonstrate the advantage of our method in QoS improvement.

Mobile traffic prediction plays a crucial role in enabling efficient network management and service provisioning. Traditional prediction approaches treat different mobile application services (such as Uber, Facebook, Twitter, etc) as isolated entities, neglecting potential correlation among them. Moreover, such isolated prediction methods necessitate the uploading of historical traffic data from all regions to forecast city-wide traffic, resulting in consuming substantial bandwidth resources and risking prediction failure in the event of data loss in specific regions. To address these challenges, we propose a novel Cross-service Attention-based Spatial-Temporal Graph Convolutional Network (CsASTGCN) for precise and communication-efficient multi-service mobile traffic prediction. Our methodology allows each mobile service to transmit the traffic data of only a fraction of regions for city-wide traffic prediction of all mobile services, which reduces the resource consumption caused by data transmission. Specifically, the sparse traffic data are initially transmitted to the cloud server and the masked graph autoencoder is utilized to roughly reconstruct the traffic volume for regions with missing data. Subsequently, a cross-service attention-based predictor is designed to calculate the data correlation among different mobile services within the same region. Considering the constantly emerging mobile services, we incorporate a novel model-based adaptive transfer learning scheme to extract valuable knowledge from the existing models and expedite the training of a new model for a new service without training from scratch, thereby enhancing the scalability of our framework. Extensive experiments conducted on a large-scale real-world mobile traffic dataset demonstrate that our model greatly outperforms the existing schemes, enhancing both the communication-efficiency and robustness of large-scale multi-service traffic prediction.

Multi-party Mobile Virtual Reality (MMVR) enables multiple mobile users to share virtual scenes for immersive multimedia experience in scenarios such as gaming, social interaction, and industrial mission collaboration. Dynamic 3D Point Cloud (DPCL) is an emerging representation form of MMVR that can be consumed as a free-viewpoint video with 6 degrees of freedom. Given that it is challenging to render DPCL at a satisfying frame rate with limited on-device resources, offloading rendering tasks to edge servers is recognized as a practical solution. However, repeated loading of DPCL scenes with a substantial amount of metadata introduces a significant redundancy overhead that cannot be overlooked when enabling multiple edge servers to support the rendering requirements of user groups. In this paper, we design PoClVR, an edge-assisted DPCL rendering system for MMVR applications, which breaks down the rendering process of the complete dynamic scene into multiple rendering tasks of dynamic objects. PoClVR significantly reduces the repetitive loading overhead of DPCL scenes on edge servers and periodically adjusts the rendering task allocation during the application running to accommodate rendering requirements. We deploy PoClVR based on a real-world implementation and the experimental evaluation results show that PoClVR can reduce GPU utilization by up to 15.1% and increase rendering frame rate by up to 34.6% compared to other baselines while ensuring that the image quality viewed by the user is virtually unchanged.

On-device Deep Neural Network (DNN) training has been recognized as crucial for privacy-preserving machine learning at the edge. However, the intensive training workload and limited onboard computing resources pose significant challenges to the availability and efficiency of model training. While existing works address these challenges through native resource management optimization, we instead leverage our observation that edge environments usually comprise a rich set of accompanying trusted edge devices with idle resources beyond a single terminal. We propose Asteroid, a distributed edge training system that breaks the resource walls across heterogeneous edge devices for efficient model training acceleration. Asteroid adopts a hybrid pipeline parallelism to orchestrate distributed training, along with a judicious parallelism planning for maximizing throughput under certain resource constraints. Furthermore, a fault-tolerant yet lightweight pipeline replay mechanism is developed to tame the device-level dynamics for training robustness and performance stability. We implement Asteroid on heterogeneous edge devices with both vision and language models, demonstrating up to 12.2× faster training than conventional parallelism methods and 2.1× faster than state-of-the-art hybrid parallelism methods through evaluations. Furthermore, Asteroid can recover training pipeline 14× faster than baseline methods while preserving comparable throughput despite unexpected device exiting and failure.

3D Gaussian Splatting (3DGS) has recently emerged as a prominent technique in novel view synthesis. The superior performance of 3DGS has catalyzed an increasing number of 3DGS- based applications in edge scenarios, where 3DGS is utilized for various purposes, such as scene representation, comprehension, and generation. Meanwhile, these edge applications also serve as primary sources of scene observations for producing 3DGS models. However, the intensive computation involved in 3DGS training and the massive number of 3D Gaussian primitives required for high-resolution scene repre-sentation hinder the effectiveness of in-situ 3DGS training on off-the-shelf edge devices, whether using standalone training or Data-Distributed-Parallel (DDP) training. To address this issue, this work proposes MIX3D, a novel mixed representation for communication-efficient distributed 3DGS training in edge scenarios. MIX3D features a global sparse sub-model and various local dense sub-models, where the sparse sub-model encodes coarse-grained appearance for the entire scene, and each dense sub-model targets fine-grained details for a specific region of the scene. Extensive evaluations on a four-device edge cluster demonstrate the effectiveness of our developed distributed 3DGS training workflow based on MIX3D, achieving reductions in training time up to 86.6% compared to vanilla DDP training and an average speedup of 3.767x over standalone training.

Federated learning (FL) is a collaborative machine learning approach that enables multiple clients to train models without sharing their private data. With the rise of deep learning, large-scale models have garnered significant attention due to their exceptional performance. However, a key challenge in FL is the limitation imposed by clients with constrained computational and communication resources, which hampers the deployment of these large models. The Mixture of Experts (MoE) architecture addresses this challenge with its sparse activation property, which reduces computational workload and communication demands during inference and updates. Additionally, MoE facilitates better personalization by allowing each expert to specialize in different subsets of the data distribution. To alleviate the communication burdens between the server and clients, we propose FedMoE-DA, a new FL model training framework that leverages the MoE architecture and incorporates a novel domain-aware, fine-grained aggregation strategy to enhance the robustness, personalizability, and communication efficiency simultaneously. Specifically, the correlation between both intra-client expert models and inter-client data heterogeneity is exploited. Moreover, we utilize peer-to-peer (P2P) communication between clients for selective expert model synchronization, thus significantly reducing the server-client transmissions. Experiments demonstrate that our FedMoE-DA achieves excellent performance while reducing the communication pressure on the server.

With the widespread adoption of video streaming applications, effective video delivery solutions are crucial for providing seamless user experiences. Recent studies have revealed that multipath transmissions are beneficial to video streaming applications, given their potential of better load balancing and fault tolerance, relative to single path settings. However, the necessity of cross-layer co-design of multipath routing and video frame scheduling is overlooked. This work identifies that preset or path-oblivious frame scheduling used in existing works cannot adapt to network dynamics and fail to enhance the quality of experiences (QoE) in multipath transmissions. Therefore, we propose MPVSched, a novel framework that unifies the design of multipath routing and application-layer frame scheduling, with a particular focus on improving the rebuffer rate for short video delivery. At the network layer, we propose to use network-assisted routing that selects the optimal paths for each video transmission, with per-hop per-frame latency prediction. We implement an end-to-end QUIC-based video streaming system by integrating our routing strategy and application-layer frame scheduler, which effectively improves streaming efficiency and prevents user-side freezes. Our testbed experiments with real-world short video request traces demonstrate that MPVSched can achieve reductions of up to 28.58% in rebuffer ratio, compared to representative baseline methods.

The open radio access network (O-RAN) architecture provides enhanced opportunities for integrating machine learning in 5G/6G resource management by decomposing RAN functionalities. Yet, generic learning mechanisms either do not fully exploit the disaggregated non-real-time and near-real-time RAN controllers or ignore the potential elasticity of application demands, another degree of freedom in managing RAN resources. We introduce a two-timescale framework aimed at optimizing users’ long-term total QoS. Rather than reactive resource allocation, our approach proactively modifies multi-resource user demands using congestion indicators, prior to enforcing any allocation rules. Addressing the issue of insufficient user feedback on individual resource utilities, we employ a bandit-feedback version of the combinatorial multi-armed bandit framework to deduce resource-specific signals. Also, to compensate for insufficient and infrequent feedback, we’ve developed an algorithm that gleans side information from live network traffic to refine predictions on user resource sensitivities. This streamlines the algorithm’s optimality convergence and leverages the two-tier O-RAN controller structure. We validate our algorithms’ efficacy through analysis and 5G usage experiments, revealing our proposed method improves application utility by 13-60%, throughput by 8-19%, and reduces latency by 10-18%.

Edge computing has recently emerged as a promising paradigm to boost the performance of distributed learning by leveraging the distributed resources at edge nodes. Architecturally, the introduction of edge nodes adds an additional intermediate layer between the master and workers in the original distributed learning systems, potentially leading to more severe straggler effect. Recently, coding theory-based approaches have been proposed for stragglers mitigation in distributed learning, but the majority focus on the conventional workers-master architecture. In this paper, along a different line, we investigate the problem of mitigating the straggler effect in hierarchical distributed learning systems with an additional layer composed of edge nodes. Technically, we first derive the fundamental trade-off between the computational loads of workers and the stragglers tolerance. Then, we propose a hierarchical gradient coding framework, which provides better stragglers mitigation, to achieve the derived computational trade-off. To further improve the performance of our framework in heterogeneous scenarios, we formulate an optimization problem with the objective of minimizing the expected execution time for each iteration in the learning process. We develop an efficient algorithm to mathematically solve the problem by outputting the optimum strategy. Extensive simulation results demonstrate the superiority of our schemes compared with conventional solutions.

Next point-of-interest (POI) recommendation has been a hot research topic, which enables new paradigms for kinds of location-based services in real-world scenarios. Due to the privacy concerns and rigorous data regulations, federated learning provides a distributed learning framework to collaboratively train the recommendation model without sharing the highly sensitive POI data with others. However, there exist two main challenges, namelylocation noise, andbalance between personalization and knowledge sharing, seriously restrict the development of the federated next POI recommendation. To this end, in this work, we propose an adaptive clustering based personalized federated learning framework for next POI recommendation with location noise, namedCPF-POI, to address the above challenges. In detail, within the local client, a location recovery module can efficiently remove noises under the given assumption from the noisy POI data in which the recovery error bound can be theoretically proved. Then, within the parameter server, an adaptive clustering scheme is proposed to capture the internal relatedness among all clients to augment positive knowledge sharing. In order to make a balance between personalization and knowledge sharing under personalized federated learning framework, we design an alternative optimization process between clustering similar clients and minimizing local personalized loss functions. Finally, extensive experiments are conducted on two diverse real-world datasets to show the advantages ofCPF-POIover state-of-the-art methods. improvement across all metrics on average.

With the proliferation of edge computing, many business entities deploy their own edge servers to compete for users, which forms multi-server edge computing networks. However, no prior work studies the competition among heterogeneous edge servers and how the competition affects users’ selfish computation offloading behaviors in such a network from an economic perspective. In this paper, we model the interactions between edge servers and users as a two-stage game. In Stage I, edge servers with heterogeneous marginal costs set their service prices to compete for users, and in Stage II, each user selfishly offloads its task to one of the edge servers or the remote cloud. Analyzing the equilibrium of the two-stage game is challenging due to edge servers’ heterogeneity and the congestion effect caused by resource sharing among users. We first investigate the equilibrium when edge servers follow the serve-as-arrive (SAA) model (i.e., serving all offloaded tasks simultaneously), and then extend our analysis to the serve-in-queue (SIQ) model (i.e., serving offloaded tasks one by one following the M/M/1 queue rule). Under the SAA model, we prove that users’ selfish computation offloading game in Stage II is a potential game and admits a unique Nash equilibrium (NE), for which we derive the explicit expressions. Furthermore, for edge servers’ price competition game in Stage I, we characterize the conditions for the uniqueness of the NE and derive its explicit expression. Under the SIQ model, we derive the unique NE of users’ selfish computation offloading game, and show that the NE of edge servers’ price competition game may not always exist. We compare the equilibrium under the two service models and show that at equilibrium, edge servers with low marginal costs can achieve higher profits under the SIQ model when edge servers’ computation capacity is large or the delay incurred on the cloud is moderate; however, edge servers with high marginal costs can obtain higher profits under the SAA model in most cases.

Federated Learning (FL) requires frequent exchange of model parameters, which leads to long communication delay, especially when the network environments of clients vary greatly. Moreover, the parameter server needs to wait for the slowest client (i.e., straggler, which may have the largest model size, lowest computing capability or worst network condition) to upload parameters, which may significantly degrade the communication efficiency. Commonly-used client selection methods such as partial client selection would lead to the waste of computing resources and weaken the generalization of the global model. To tackle this problem, along a different line, in this paper, we advocate the approach of model parameter dropout instead of client selection, and accordingly propose a novel framework of Federated learning scheme with Differential parameter Dropout (FedDD). FedDD consists of two key modules: dropout rate allocation and uploaded parameter selection, which will optimize the model parameter uploading ratios tailored to different clients' heterogeneous conditions and also select the proper set of important model parameters for uploading subject to clients' dropout rate constraints. Specifically, the dropout rate allocation is formulated as a convex optimization problem, taking system heterogeneity, data heterogeneity, and model heterogeneity among clients into consideration. The uploaded parameter selection strategy prioritizes on eliciting important parameters for uploading to speedup convergence. Furthermore, we theoretically analyze the convergence of the proposed FedDD scheme. Extensive performance evaluations demonstrate that the proposed FedDD scheme can achieve outstanding performances in both communication efficiency and model convergence, and also possesses a strong generalization capability to data of rare classes.

Given that devices (i.e., clients) participating in federated edge learning (FEL) are autonomous and resource-constrained in nature, it is critical to design effective incentive mechanisms to encourage client participation so as to improve the performance of FEL. In this article, we aim to boost the FEL training efficiency by answering how much compute resource should clients autonomously contribute to maximize their utilities. To this end, we develop AutoFL, an autonomous client participation decision framework for federated learning at the network edge without assuming that each client possesses complete information. We first model the problem of autonomous client participation as a Bayesian game with incomplete information, where each player in the game is associated with a set of types according to network conditions. We optimize an individual client's decision based on the dynamics of the population estimated following the Bayes rule. We prove that AutoFL can converge to a unique Bayesian Nash equilibrium point. Empirical results on three real datasets show that AutoFL achieves a higher model accuracy with only 15.5-24.5% model aggregation time per global training round, and its energy cost saving on mobile devices is 82.2-86.8% compared to the state-of-the-art algorithms. Moreover, we can achieve a 2.75-3.2x long-term fairness compared to classical solutions.

Federated learning (FL) has emerged as a promising paradigm that enables clients to collaboratively train a shared global model without uploading their local data. To alleviate the heterogeneous data quality among clients, artificial intelligence-generated content (AIGC) can be leveraged as a novel data synthesis technique for FL model performance enhancement. Due to various costs incurred by AIGC-empowered FL (e.g., costs of local model computation and data synthesis), however, clients are usually reluctant to participate in FL without adequate economic incentives, which leads to an unexplored critical issue for enabling AIGC-empowered FL. To fill this gap, we first devise a data quality assessment method for data samples generated by AIGC and rigorously analyze the convergence performance of FL model trained using a blend of authentic and AI-generated data samples. We then propose a data quality-aware incentive mechanism to encourage clients’ participation. In light of information asymmetry incurred by clients’ private multi-dimensional attributes, we investigate clients’ behavior patterns and derive the server's optimal incentive strategies to minimize server's cost in terms of both model accuracy loss and incentive payments for both complete and incomplete information scenarios. Numerical results demonstrate that our proposed mechanism exhibits highest training accuracy and reduces up to 53.34% of the server's cost with real-world datasets, compared with existing benchmark mechanisms.

Within the paradigm of forthcoming 6G network infrastructures, unmanned aerial vehicles (UAVs), functioning as principal conveyances, are projected to emerge as pivotal enablers in the nascent domain of the low-altitude economy. UAVs are poised to embrace various innovative applications, including latency-sensitive and compute-intensive services. However, UAVs are constrained by their energy capacity and computational resources, rendering them insufficient for fulfilling the increasingly rigorous service demands in the future. To address these challenges, our investigation focuses on the innovative UAV-based Vehicular Edge Computing (UVEC) framework, incorporating Vehicular Edge Computing (VEC) in UAV systems to bolster service reliability. A UAV can enhance its mission duration by dynamically selecting suitable vehicles for computation offloading and adaptively adjusting the task offloading ratio between vehicles and the edge server. By integrating vehicle selection and task offloading scheduling in the UVEC framework, we investigate the optimization of energy efficiency while satisfying the statistical delay and the buffer constraints for UAVs. To deal with the proposed problem, a distributed algorithm is designed by jointly considering the vehicle selection for task offloading radio to vehicles and the edge server. The stochastic network calculus (SNC) is employed to derive performance bounds for the statistical delay and constraints, enabling robust analysis and optimization of network performance. After that, we leverage linear transformation techniques to reformulate the original problem into a linear framework, enabling the application of the Alternating Direction Method of Multipliers (ADMM) algorithm to efficiently solve the transformed problem. Theoretical analysis and simulation results show that our algorithm converges while effectively satisfying service reliability constraints within the desired targets, outperforming benchmark schemes in terms of efficiency while meeting task delay and error-rate bounded constraints.

Deep learning classifiers are crucial in the age of artificial intelligence. The device-edge-based collaborative inference has been widely adopted as an efficient framework for promoting its applications in IoT and 5G/6G networks. However, it suffers from accuracy degradation under non-i.i.d. data distribution and privacy disclosure. For accuracy degradation, direct use of transfer learning and split learning is high cost and privacy issues remain. For privacy disclosure, cryptography-based approaches lead to a huge overhead. Other lightweight methods assume that the ground truth is non-sensitive and can be exposed. But for many applications, the ground truth is the user's crucial privacy-sensitive information. In this paper, we propose a framework of Roulette, which is a task-oriented semantic privacy-preserving collaborative inference framework for deep learning classifiers. More than input data, we treat the ground truth of the data as private information. We develop a novel paradigm of split learning where the back-end DNN is frozen and the front-end DNN is retrained to be both a feature extractor and an encryptor. Moreover, we provide a differential privacy guarantee and analyze the hardness of ground truth inference attacks. To validate the proposed Roulette, we conduct extensive performance evaluations using realistic datasets, which demonstrate that Roulette can effectively defend against various attacks and meanwhile achieve good model accuracy. In a situation where the non-i.i.d. is very severe, Roulette improves the inference accuracy by 21% averaged over benchmarks, while making the accuracy of discrimination attacks almost equivalent to random guessing.

Federated learning (FL) provides a collaborative paradigm for distributedly training a global model while protecting clients' privacy. In addition to communication bottlenecks and non-i.i.d. data distributions, the FL framework introduces two fundamental economic challenges: first, clients are self-interested and strategic in practice, requiring specific incentives to participate in FL; second, each client can misreport its private information to its advantage. Although existing studies have proposed economic mechanisms, they are often restricted to a “binary” participation scenario, leading to communication overheads or biased models due to client heterogeneity. In this paper, we first analyze the convergence bound under arbitrary client sampling probability with a varying number of clients. Then, we consider an optimal mechanism design problem: the FL convergence bound minimization subject to budget constraint, incentive compatibility, and individual rationality. We derive the optimal sampling probability function in a close form. To overcome the unknown prior distribution challenge, we introduce a prior-independent mechanism design, and show how it gradually learns cost distributions by exploiting the incentive compatibility property. We perform extensive experiments and show that, while outperforming the uniform sampling scheme, two proposed schemes (prior-based and prior-independent ones) perform closely to the ideal complete information upper bound.

Deep learning (DL) powered real-time applications usually need continuous training using data streams generated over time and across different geographical locations. Enabling data offloading among computation nodes through model training is promising to mitigate the problem that devices generating large datasets may have low computation capability. However, offloading can compromise model convergence and incur communication costs, which must be balanced with the long-term cost spent on computation and model synchronization. Therefore, this paper proposes EdgeC3, a novel framework that can optimize the frequency of model aggregation and dynamic offloading for continuously generated data streams, navigating the trade-off between long-term accuracy and cost. We first provide a new error bound to capture the impacts of data dynamics that are varying over time and heterogeneous across devices, as well as quantifying varied data heterogeneity between local models and the global one. Based on the bound, we design a two-timescale online optimization framework. We periodically learn the synchronization frequency to adapt with uncertain future offloading and network changes. In the finer timescale, we manage online offloading by extending Lyapunov optimization techniques to handle an unconventional setting, where our long-term global constraint can have abruptly changed aggregation frequencies that are decided in the longer timescale. Finally, we theoretically prove the convergence of EdgeC3 by integrating the coupled effects of our two-timescale decisions, and we demonstrate its advantage through extensive experiments performing distributed DL training for different domains.

As the primary driver of intelligent mobile applications, deep neural networks (DNNs) have gradually deployed to millions of mobile devices, producing massive latency-sensitive and computation-intensive tasks daily. Mobile edge computing facilitates the deployment of computing resources at the edge, which enables fine-grained offloading of DNN inference tasks from mobile devices to edge nodes. However, most existing studies have not systematically considered three crucial performance aspects: scheduling multiple streams of DNN inference tasks, leveraging multi-exit models to hasten task processing, and partitioning inference models for partial offloading. To this end, this paper proposes an adaptive inference framework in mobile edge computing, which can dynamically select the exit point and partition point for multiple inference task streams. We design a dynamic programming algorithm to obtain an efficient solution under the ideal condition that task arrival information is known. Further, we design a learning-based algorithm for online scheduling, whose training efficiency is improved based on historical experience initialization and priority experience replay. Experimental results show that compared with the Greedy algorithm, the online algorithm improves the performance on two environmental parameters by an average of 5.9% and 32%, respectively.

Federated Learning (FL) is a distributed learning paradigm that can coordinate heterogeneous edge devices to perform model training without sharing private data. While prior works have focused on analyzing FL convergence with respect to hyperparameters like batch size and aggregation frequency, the joint effects of adjusting these parameters on model performance, training time, and resource consumption have been overlooked, especially when facing dynamic data streams and network characteristics. This paper introduces novel analytical models and optimization algorithms that leverage the interplay between batch size and aggregation frequency to navigate the trade-offs among convergence, cost, and completion time for dynamic FL training. We establish a new convergence bound for training error considering heterogeneous datasets across devices and derive closed-form solutions for co-optimized batch size and aggregation frequency that are consistent across all devices. Additionally, we design an efficient algorithm for assigning different batch configurations across devices, improving model accuracy and addressing the heterogeneity of both data and system characteristics. Further, we propose an adaptive control algorithm that dynamically estimates network states, efficiently samples appropriate data batches, and effectively adjusts batch sizes and aggregation frequency on the fly. Extensive experiments demonstrate the superiority of our offline optimal solutions and online adaptive algorithm.

Cross-silo federated learning (FL) is a distributed learning method where clients collaboratively train a global model without exchanging local data. However, recent works reveal that potential privacy leakage occurs when clients upload their local updates. Although some works have studied privacy-preserving mechanisms in FL, the selfish privacy-preserving behaviors of clients (who are usually cost-sensitive companies or organizations) are yet to be explored. In this paper, we formulate clients' privacy-preserving behaviors in cross-silo FL as a multi-stage privacy preservation game, where each stage game corresponds to one training iteration. Specifically, clients selfishly perturb their local updates in each training iteration to trade off between convergence performance and privacy loss. To analyze the game, we first derive a novel theoretical bound to characterize the impact of clients' local perturbations on the convergence of FL through analyzing the corrective effect of gradient descent in model training. With the novel convergence bound, we prove that each stage game is a potential game with a unique Nash equilibrium (NE) and the multi-stage privacy preservation game admits a unique subgame perfect Nash equilibrium (SPNE). We show that at the SPNE, the magnitude of each client's local perturbation decreases geometrically with training iterations. We then characterize the efficiency of the SPNE in terms of social cost by the price of anarchy (PoA), and show that the efficiency decreases with the number of clients in some cases. To tackle this problem, we propose a socially efficient incentive mechanism that allows monetary transfer among clients and guarantees individual rationality, budget balance, and social efficiency. To further elicit the private information from the selfish clients, we propose a truthful mechanism that achieves approximate social efficiency. Simulation results show that our proposed mechanisms are effective even when clients are highly heterogeneous, and can decrease clients' total cost by up to 58.08% compared with that at the SPNE.

Ultra-High-Definition (UHD) videos have been getting increasing attention. However, existing video streaming solutions fail to deliver them due to the extremely high bandwidth requirement. The emerging cloud native 5G networks have opened up the possibility of enhancing UHD video quality by leveraging in-network video streaming. Unfortunately, the restricted storage and bandwidth of in-network servers could become the main bottleneck. To this end, we present${\sf EMS}$, a novel UHD video streaming framework, by integratingErasure-coded storage withMulti-sourceStreaming. We respectively introduce a deadline-aware and a latency-sensitive metric to indicate the service quality of video servers and advocate a federated learning paradigm for the adaptive service quality update, including a reinforcement learning based multi-server selection (i.e., user local training) and a global service quality aggregation. To facilitate user local training without sacrificing streaming Quality-of-Experience (QoE), we cast the multi-server selection associated with the restriction on the average number of selected servers per video chunk into two kinds of Multi-Armed Bandit (MAB) models in terms of the proposed service quality metrics. We design lightweight Upper Confidence Bound (UCB) based algorithms with a theoretical performance guarantee. We implement a prototype of${\sf EMS}$, and extensive experiments confirm the superiority of the proposed algorithms.

Applying the serverless paradigm to edge computing improves edge resource utilization while bringing the benefits of flexible scaling and pay-as-you-go to latency-sensitive applications. This extends the boundaries of serverless computing and improves the quality of service for Function-as-a-Service users. However, as an emerging cloud computing paradigm, serverless edge computing faces pressing challenges, with one of the biggest obstacles being delay caused by excessively long container cold starts. Cold start delay is defined as the time between when a serverless function is triggered and when it begins to execute, and its existence seriously impacts resource utilization and Quality of Service (QoS). In this paper, we study how to minimize the total system cost by caching function containers and selecting routes for neighboring functions via edge or public clouds. We prove that the proposed problem is NP-hard even in the special case where the user request contains only one function, and that the unpredictability of user requests and the impact between adjacent time decisions require that the problem to be solved in an online fashion. We then design the Online Lazy Caching algorithm, an online algorithm with a worst-case competitive ratio using a randomized dependent rounding algorithm to solve the problem. Extensive simulation results show that the proposed online algorithm can achieve close-to-optimal performance in terms of both total cost and cold start cost compared to other existing algorithms, with average improvements of 31.6% and 51.7%.

Low Earth orbit (LEO) satellites enable complex Earth observation tasks (e.g.,remote sensing and cooperative monitoring) by leveraging large-scale satellite-generated Earth imageries and state-of-the-art machine learning (ML) techniques. However, due to restricted downlink bandwidth and spotty connectivity, it is infeasible for the satellites to transmit all the imageries to ground stations for ML model training. To address this issue, we use federated learning (FL) to mitigate the significant overhead of raw data transmission only by enabling model parameter exchange. Traditional FL requires a central server for model parameter aggregation, which is impractical for distributed LEO satellite constellation due to the difficulty of identifying a suitable central satellite. To tackle such challenge, we take the unique topological characteristics of the LEO satellite constellation to design a decentralized FL framework that enables efficient model aggregation in LEO satellite networks without a central server. The framework can avoid the reliability and communication bandwidth problems of the central server in centralized FL. To mitigate the straggler effect and address the statistical heterogeneity, we then propose a novel offloading framework for decentralized FL in LEO satellite networks to aid the collaboration among multiple satellites for resource sharing. Based on it, we derive a satellite-centric threshold-based offloading strategy and a system-wide greedy-based iterative offloading decision making algorithm, in order to achieve delay and accuracy optimization under the computation and communication power constraints. Theoretical analysis demonstrates that the proposed framework contributes to the high training performance of the global model. Extensive experiments based on realistic datasets show that the proposed framework can reduce the system delay by up to 41% on average and improve the global model accuracy by up to 9.39% compared with benchmark policies.

The model accuracy achieved by federated learning (FL) depends significantly on devices' data distributions. To improve the model accuracy of FL with heterogeneous data distributions on devices, existing works propose some device sampling methods for the central server, but face the problem that the selected devices may still have unbalanced data. In this paper, we propose a novel coalitional FL framework for FL with heterogeneous data. Specifically, devices can cooperate and form device coalitions to reduce the data unbalancedness, and we formulate devices' interactions as a coalition formation game. Then the server selects an optimal subset of device coalitions to improve the model accuracy. Analyzing the coalition formation and selection framework is challenging since the relationship between model accuracy and data heterogeneity is not clear, and devices' coalition formation decisions and the server's coalition selection strategy are coupled in a highly non-trivial manner. We first derive a novel theoretical characterization of the relationship between model accuracy loss and data heterogeneity which follows an inverse function. With the novel theoretical relationship, we analyze devices' coalition formation game. We characterize the conditions under which the Nash stable partition exists, and propose an accelerated algorithm for devices to reach the Nash stable partition. For the server's device coalition selection problem, we show that the model accuracy loss depends on both data heterogeneity and the number of data samples of device coalitions in a non-monotonous way, and we propose a low-complexity algorithm for the server to select device coalitions efficiently. We conduct extensive simulations and show that our proposed coalition formation and selection framework reduces the data heterogeneity of selected device coalitions by up to$58.6\%$and increases the model accuracy by up to$6.8\%$compared with four existing benchmarks.

Serverless computing is envisioned as the de-facto standard for next-generation cloud computing. However, the cold start dilemma has impeded its adoption by delay-sensitive and burst applications. In this paper, we propose to tame serverless cold start in a cloud inference system with edge computing. Specifically, the proposed solution smooths the serverless cloud workload with user-owned edge computing, reducing the number of cold starts. Leveraging the configurability of requests and serverless functions, the proposed solution further reduces the transmission latency and serverless cost by adapting request configuration (e.g., image resolution) and function configuration (e.g., memory). To alleviate the potential inference accuracy degradation incurred by configuration adaption, we aim to strike a nice balance between inference latency, cost, and accuracy. However, achieving this goal is non-trivial since the underlying optimization is non-convex and involves future uncertain information. To simultaneously address dual challenges, the presented cold-start-aware online algorithms apply the regularization technique to decompose the problem into separate convex subproblems. Then, it applies lazy switching to smooth the number of provisioned functions and thus reduces the cold start. Through rigorous theoretical analysis, realistic prototype evaluations on AWS Lambda, and trace-driven simulations, we comprehensively validate the theoretical and empirical performance of our proposed solution.

Collaborative Edge Computing (CEC) is an emerging paradigm that collaborates heterogeneous edge devices as a resource pool to compute DNN inference tasks in proximity such as edge video analytics. Nevertheless, as the key knob to improve network utility in CEC, existing works mainly focus on the workload routing strategies among edge devices with the aim of minimizing the routing cost, remaining an open question for joint workload allocation and routing optimization problem from a system perspective. To this end, this paper presents a holistic, learned optimization for CEC towards maximizing the total network utility in an online manner, even though the utility functions of task input rates are unknown a priori. In particular, we characterize the CEC system in a flow model and formulate an online learning problem in a form of cross-layer optimization. We propose a nested-loop algorithm to solve workload allocation and distributed routing iteratively, using the tools of gradient sampling and online mirror descent. To improve the convergence rate over the nested-loop version, we further devise a single-loop algorithm. Rigorous analysis is provided to show its inherent convexity, efficient convergence, as well as algorithmic optimality. Finally, extensive numerical simulations demonstrate the superior performance of our solutions.

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.

Federated learning (FL) enables multiple data owners to collaboratively train machine learning (ML) models for different model requesters while keeping data localized. Thus, FL can mitigate privacy leakage in conventional data marketplaces for ML applications requiring raw data trading for centralized model training. Nevertheless, data owners involved in FL may still suffer potential privacy leakage from gradient exposure to the model requesters. In this work, we advocate a novel data marketplace with differentially private federated learning (DPFL) to reduce such threats and maximize the social welfare. Designing such a marketplace involves several challenges. First, it is difficult to determine the privacy budget that a data owner should choose for a model requester, since they have conflicting objectives and private utility/cost information. Second, each data owner sustains privacy costs from his friends’ participation in DPFL due to data correlations, which introduces a negative externality to the market. We design a social-aware iterative double auction (SARDA) mechanism to resolve these challenges and achieve socially optimal market operation. SARDA employs a broker to coordinate the interactions between data owners and model requesters and induces them to truthfully report by iteratively updating the allocation and pricing rules. Moreover, SARDA accounts for the negative externality by incorporating others’ bids to reimburse each data owner. We show that SARDA achieves the optimal social performance and creates up to$60\%$higher social welfare than the social-agnostic benchmark.

Accurate navigation is of paramount importance to ensure flight safety and efficiency for autonomous drones. Recent research starts to use Deep Neural Networks (DNN) to enhance drone navigation given their remarkable predictive capability for visual perception. However, existing solutions either run DNN inference tasks on drones in situ, impeded by the limited onboard resource, or offload the computation to external servers which may incur large network latency. Few works consider jointly optimizing the offloading decisions along with image transmission configurations and adapting them on the fly. In this paper, we propose A3D, an edge server assisted drone navigation framework that can dynamically adjust task execution location, input resolution, and image compression ratio in order to achieve low inference latency, high prediction accuracy, and long flight distances. Specifically, we first augment state-of-the-art convolutional neural networks for drone navigation and define a novel metric called Quality of Navigation as our optimization objective which can effectively capture the above goals. We then design a deep reinforcement learning (DRL) based neural scheduler at the drone side for which an information encoder is devised to reshape the state features and thus improve its learning ability. To further support simultaneous multi-drone serving, we extend the edge server design by developing a network-aware resource allocation algorithm, which allows provisioning containerized resources aligned with drones’ demand. We finally implement a proof-of-concept prototype with realistic devices and validate its performance in a real-world campus scene, as well as a simulation environment for thorough evaluation upon AirSim. Extensive experimental results show that A3D can reduce end-to-end latency by 28.06% and extend the flight distance by up to 27.28% compared with non-adaptive solutions.

Graph Neural Networks (GNNs) have gained growing interest in miscellaneous applications owing to their outstanding ability in extracting latent representation on graph structures. To render GNN-based service for IoT-driven smart applications, traditional model serving paradigms usually resort to the cloud by fully uploading geo-distributed input data to remote datacenters. However, our empirical measurements reveal the significant communication overhead of such cloud-based serving and highlight the profound potential in applying the emerging fog computing. To maximize the architectural benefits brought by fog computing, in this paper, we present Fograph, a novel distributed real-time GNN inference framework that leverages diverse and dynamic resources of multiple fog nodes in proximity to IoT data sources. By introducing heterogeneity-aware execution planning and GNN-specific compression techniques, Fograph tailors its design to well accommodate the unique characteristics of GNN serving in fog environments. Prototype-based evaluation and case study demonstrate that Fograph significantly outperforms the state-of-the-art cloud serving and fog deployment by up to 5.39$\times$execution speedup and 6.84$\times$throughput improvement.

Nowadays, the emerging short video streaming applications have gained substantial attention. With the rapidly burgeoning demand for short video streaming services, maximizing their Quality of Experience (QoE) is an onerous challenge. Current video preloading algorithms cannot determine video preloading sequence decisions appropriately due to the impact of users’ swipes and bandwidth fluctuations. As a result, it is still ambiguous how to improve the overall QoE while mitigating bandwidth wastage to optimize short video streaming services. In this article, we devise Gamora, a buffer-aware short video streaming system to provide a high QoE of users. In Gamora, we first propose an unordered preloading algorithm that utilizes a Deep Reinforcement Learning (DRL) algorithm to make video preloading decisions. Then, we further devise an Asymmetric Imitation Learning (AIL) algorithm to guide the DRL-based preloading algorithm, which enables the agent to learn from expert demonstrations for fast convergence. Finally, we implement our proposed short video streaming system prototype and evaluate the performance of Gamora on various real-world network datasets. Our results demonstrate that Gamora significantly achieves QoE improvement by 28.7%–51.4% compared to state-of-the-art algorithms, while mitigating bandwidth wastage by 40.7%–83.2% without sacrificing video quality.

As a privacy-preserving distributed learning paradigm, federated learning (FL) enables multiple client devices to train a shared model without uploading their local data. To further enhance the privacy protection performance of FL, differential privacy (DP) has been successfully incorporated into FL systems to defend against privacy attacks from adversaries. In FL with DP, how to stimulate efficient client collaboration is vital for the FL server due to the privacy-preserving nature of DP and the heterogeneity of various costs (e.g., computation cost) of the participating clients. However, this kind of collaboration remains largely unexplored in existing works. To fill in this gap, we propose a novel analytical framework based on Stackelberg game to model the collaboration behaviors among clients and the server with reward allocation as incentive in FL with DP. We first conduct rigorous convergence analysis of FL with DP and reveal how clients’ multidimensional attributes would affect the convergence performance of FL model. Accordingly, we solve the Stackelberg game and derive the collaboration strategies for both clients and the server. We further devise an approximately optimal algorithm for the server to efficiently conduct the joint optimization of the client set selection, the number of global iterations, and the reward payment for the clients. Numerical evaluations using real-world datasets validate our theoretical analysis and corroborate the superior performance of the proposed solution.

Recent advances in deep reinforcement learning (DRL) have made it possible to train various powerful agents to perform complex tasks in real-time environments. With the next-generation communication technologies, making cloud-edge collaborative artificial intelligence service with evolved DRL agents can be a significant scenario. However, agents with different algorithms and architectures in the same DRL scenario may not be compatible, and training them is either time-consuming or resource-demanding. In this paper, we design a novel cloud-edge collaborative DRL training framework, named Offline-Transfer-Online, which is a new approach that can speed up the convergence of online DRL agents at the edge by interacting with offline agents in the cloud, with the minimum data interchanged and without relying on high-quality offline datasets. Therein, we propose a novel algorithm-independent knowledge distillation algorithm for online RL agents, by leveraging pre-trained models and the interface between agents and the environment to transfer distilled knowledge among multiple heterogeneous agents efficiently. Extensive experiments show that our algorithm can accelerate the convergence of various online agents in a double to decuple speed, with comparable reward achieved in different environments.

Video analytics play a pivotal role in enhancing the safety of intelligent surveillance and autonomous driving. However, the transmission of vast video data and the computational demands of video analytics present challenges within traditional cloud computing paradigms. To address latency concerns, dynamic video analytics often leverage edge deployments. Nevertheless, the efficient allocation of resources at the edge, balancing cost-effectiveness and accuracy, becomes crucial, especially when multiple video analytics services concurrently operate within the system. This paper introduces an edge-centric incentive mechanism designed to encourage greater participation from edge nodes in offloading tasks. The key focus is on addressing the dynamic nature of edge resources and optimizing system returns through a rational pricing mechanism. We propose a decentralized Soft Actor-Critic algorithm grounded in game theory (DSACG) to autonomously learn the optimal pricing strategy. A comprehensive theoretical analysis, supported by extensive simulations, substantiates the effectiveness of our proposed solution.

The 360-degree video streaming service provides a unique immersive viewing experience for users, who can freely change their Field-of-View (FoV) to view different portions of the videos. However, the demands for high throughput and low latency for 360-degree video pose substantial challenges to the current network infrastructure. Super Resolution (SR) is the procedure for reconstructing high-resolution images from low-resolution ones. Hence, caching video content on the network edge in advance, which is near end users, and applying the SR technique can significantly alleviate the transmission latency. In this article, we describeNOVA, an efficientNeural-OptimizedViewportAdaptive 360-degree video streaming system to improve the Quality of Experience (QoE) of users. In NOVA, we first design a foveated rendering SR approach to super-resolve video tiles utilizing computational resources at the edge. Subsequently, we present a meta-learning-based Multi-Agent Reinforcement Learning (MARL) algorithm to select SR depths and video tiles inside users’ viewports for agile video tile adaptation to optimize overall QoE under frequent network fluctuations. Finally, we implement the holistic prototype of NOVA and evaluate its performance on various real-world network datasets. Extensive experiments illustrate that compared to the state-of-the-art algorithms, NOVA improves average user-perceived QoE by up to 27%.

To meet the stringent requirement of edge intelligence applications, resource-constrained devices can offload their task to nearby resource-rich devices. Resource awareness, as a prime prerequisite for offloading decision-making, is critical for achieving efficient collaborative computation performance. Although major works have explored computation offloading in dynamic edge environments, the impact of fresh resource information perception has not been formally investigated. To bridge the gap, we design a cost-aware edge resource probing (CERP) framework for infrastructure-free edge computing, where a task device self-organizes its resource probing to enable informed computation offloading. We first formulate the joint optimization of device probing and offloading as a multi-stage optimal stopping problem and derive a multi-threshold-based optimal strategy with theoretical guarantees. Accordingly, we devise a data-driven layered learning mechanism to handle more complex real-world scenarios. The layered learning enables the task device to adaptively learn the optimal probing sequence and decision thresholds on the fly, aiming to strike a good balance between the gain of choosing the best edge device and the accumulated cost of deep resource probing. To further boost its learning efficiency, we replace the$\epsilon$-greedy method with a tailored UCB-based adaptive exploration scheme in layered learning, thus better navigating the exploration and exploitation trade-off during probing processes. Finally, we conduct a thorough performance evaluation of the proposed CERP schemes using both extensive numerical simulations and realistic system prototype implementation, which demonstrate the superior performance of CERP in diverse application scenarios.

Federated learning (FL) is a distributed machine learning scheme in which clients jointly train a model without exposing their private data to a central server. However, two challenges exist: one technical challenge of the non-IID issue and one economic challenge of the incentive issue. Many existing works presented incentive mechanisms to select clients with high-quality data to tackle the non-IID issue. However, the existing works assumed the server's availability of clients' true data quality information. We notice that this assumption is hard to satisfy due to the private nature of the information. In this paper, we try to eliminate this assumption and adopt a local differentially private mechanism in the incentive mechanism. In this regard, we propose a Bayesian-based method for the server to estimate the clients' qualities and an efficient algorithm that incentivizes clients with approximately high-quality data. We prove that our solution has an approximation guarantee and is incentive-compatible, individually rational, and computationally efficient. We also analyze the quality loss due to the integration of the privacy-preserving mechanism. We conduct extensive experiments and show that our proposed solution outperforms the mechanism without considering the non-IID issue and is comparable to the mechanism without privacy protection.

With the proliferation of 5G networks, mobile edge computing (MEC) has emerged as a promising technology to fulfill the stringent requirements for reliability-aware services in the Internet of Things (IoT). However, in such networks, the wireless channel states and the task arrivals are stochastic and hard to predict well. Under this scenario, tasks generated from mobile devices would pile up in the transmission queue and edge computing queue when offloading to the edge cloud via a 5G network, resulting in quality degradation for reliability-aware services. To tackle the above challenges, we introduce non-orthogonal multiple access (NOMA) in MEC to meet the requirements of ultra-reliable and low-latency communications (URLLC), in which task queuing delay violation probability and transmission error probability are both considered. Furthermore, we explore the closed-form expression based on the effective capacity (EC) to derive the performance boundary of service reliability under a general model that multiple data sources are from different IoT devices and tasks are offloaded through two-stage transmission-computing tandem queues. Based on the above mathematical analysis for service reliability, we propose an efficient strategy combining power allocation and task offloading to reduce energy consumption for all devices in NOMA-enabled MEC. Extensive simulation studies are further conducted to validate the advantage of our strategy and show the significant performance gain of nearly up to 20% over other alternatives.

Grant-free random access is an effective technology for enabling low-overhead and low-latency massive access, where joint activity detection and channel estimation (JADCE) is a critical issue. Although existing compressed sensing algorithms can be applied for JADCE, they usually fail to simultaneously harvest the following properties: effective sparsity inducing, fast convergence, robust to different pilot sequences, and adaptive to time-varying networks. To this end, we propose an unfolding framework for JADCE based on the proximal gradient method. Specifically, we formulate the JADCE problem as a group-row-sparse matrix recovery problem and leverage a minimax concave penalty rather than the widely-used$\ell _{1}$-norm to induce sparsity. We then develop a proximal gradient-based unfolding neural network that parameterizes the algorithmic iterations. To improve convergence rate, we incorporate momentum into the unfolding neural network, and prove the accelerated convergence theoretically. Based on the convergence analysis, we further develop an adaptive-tuning algorithm, which adjusts its parameters to different signal-to-noise ratio settings. Simulations show that the proposed unfolding neural network achieves better recovery performance, convergence rate, and adaptivity than current baselines.

No abstract available.

To identify dense and small-size pedestrians in surveillance systems, high-resolution cameras are widely deployed, where high-resolution images are captured and delivered to off-the-shelf pedestrian detection models. However, given the highly computation-intensive workload brought by the high resolution, the resource-constrained cameras fail to afford accurate inference in real time. To address that, we propose Hode, an offloaded video analytic framework that utilizes multiple edge nodes in proximity to expedite pedestrian detection with high-resolution inputs. Specifically, Hode can intelligently split high-resolution images into respective regions and then offload them to distributed edge nodes to perform pedestrian detection in parallel. A spatio-temporal flow filtering method is designed to enable context-aware region partitioning, as well as a DRL-based scheduling algorithm to allow accuracy-aware load balance among heterogeneous edge nodes. Extensive evaluation results using realistic prototypes show that Hode can achieve up to 2.01× speedup with very mild accuracy loss.

Dual-functional radar-communication (DFRC) attracts extensive attention recently, given its potential to integrate the sensing and communication processes for enhancing the spectrum efficiency and hardware utilization. Due to the co-channel interference, effective resource allocation is a critical issue for DFRC, which typically relies on the accurate channel estimation. However, the conventional estimate-then-optimize algorithms may not work well due to inaccurate channel estimation, high computation complexity, and inconsistent optimization goals. This paper considers a DRFC system with multiuser multiple-input-multiple-output (MIMO) communications and MIMO radar sensing, where an end-to-end learning algorithm is developed to tackle the aforementioned issues. We formulate an optimization problem to maximize the communication performance subject to the radar sensing constraints, via optimizing both the transmit and receive beamforming matrices, while considering channel estimation in the loop. To tackle this challenging problem, we exploit the universal approximation property of the neural network to develop an end-to-end learning algorithm to directly learn the mapping between the pilot signals and the beamforming matrices, and meanwhile appropriately design the loss function to account for the radar sensing constraints. Simulations show that our proposed algorithm achieves a much greater communication performance than the baseline algorithm, while guaranteeing the same sensing performance.

Serverless computing is becoming prevalent to develop resource-demanding and delay-sensitive edge native applications across the edge and cloud. The unique pricing mechanism of serverless computing brings new opportunities to reduce the cost of edge native applications, by orchestrating function fusion and placement across the edge and cloud. However, function fusion potentially increases the latency of the serverless workflow. To navigate this performance-cost tradeoff, we present an online predictive function orchestration framework which leverages predictions to dynamically optimize the function fusion and placement. Preliminary evaluation results verify the efficacy of the proposed framework.

Despite the popularity of Serverless computing, there are insufficient efforts dedicated to Serverless workflows (i.e., Serverless function orchestration), particularly for Serverless edge computing. In this paper, we first identify the challenges of deploying the state-of-the-art cloud-oriented Serverless workflow scheduling on resource-constrained edge devices, then propose to model Serverless workflows with behavior trees, and finally reveal our key observations and preliminary results for behavior tree-based Serverless workflow scheduling.

Edge intelligence is an emerging paradigm that leverages edge computing to pave the last mile delivery of artificial intelligence. While pilot efforts on edge intelligence have mostly focused on the performance and power issues, the sustainability dilemma along with the upcoming carbon peaking and neutrality era has largely been overlooked. To green edge intelligence, we propose a carbon-aware online control framework (CARE) in this paper. CARE colocates learning and inference tasks within an edge node and dynamically adapts their configurations based on the temporal variation of carbon intensity and renewable energy availability. With such a colocation setup, CARE aims to minimize the long-term inference accuracy loss under the long-term carbon emission cap. The underlying long-term optimization problem is nontrivial since it involves uncertain information (e.g., renewable energy availability) and is NP-hard. To address these dual challenges, CARE first designs an online learning module to make fractional decisions by learning from previous system dynamics and configuration adaptation results. Then, CARE further designs a randomized rounding module, which converts the fractional decision into integer without violating the long-term carbon emission cap. The effectiveness of CARE is verified by rigorous theoretical analysis and extensive trace-driven simulations.

Edge intelligence is an emerging paradigm that leverages edge computing to pave the last-mile delivery of artificial intelligence (AI). To adapt to the resource restriction, model selection which adaptively selects DNN model variants is widely applied to shape the resource demand of edge AI inference tasks. Unfortunately, in current edge AI serving systems, applications are suffering unfairness since the DNN model selection is performed in a best-effort manner to maximize the system-wide inference accuracy. To achieve a predictable inference accuracy for the applications, edge AI serving systems should guarantee the minimum inference accuracy in a fair fashion at the application level. At the same time, edge resources should be efficiently utilized to minimize operational costs. In this paper, we model the edge DNN model selection problem as a Nash Bargaining Game (NBG), and propose the model selection principles by guaranteeing a base accuracy for each application. Based on the rigorous cooperative game-theoretic approach, we design an approximate algorithm to achieve computationally-efficient and fair model selection, corresponding to the Nash Bargaining Solution (NBS). With extensive trace-driven simulations, we show that our strategy can meet two desirable requirements towards the predictable inference accuracy for applications as well as low operational costs for the system.

Video analytics services are extensively employed in various real-time applications, including crime monitoring, business intelligence, and traffic flow control. These applications typically depend on cloud data centers to aggregate video stream tasks using Infrastructure as a Service (IaaS). However, the conventional approach of employing limited-term leased virtual machines (VMs) often leads to leads to high costs and idle time. Serverless computing or Function as a Service (FaaS) offers a more adaptable, pay-as-you-go solution but at a higher price, introducing latency challenges. Edge servers can reduce latency and costs, but limited physical resources affect accuracy. Opting for high-quality analytic services with better frame rates and resolutions can maintain accuracy but may not control costs. Therefore, the key challenge in video analytics is choosing the right configuration and computing methods for high accuracy, low cost, and low latency in edge, IaaS and FaaS environments. To address the challenge, we consider a scenario where the video analytics task is offloaded by combining the above three placements (i.e., VM, serverless, and edge node) to optimize the cost and latency while keeping the accuracy. The optimization problem can be formulated as an Integer Programming (IP) problem of which the NP-hardness is proved. To further deal with it, we propose an efficient online algorithm, which can optimize cost and latency using the Lyapunov optimization analysis while ensuring accuracy over long periods. Finally, the multi-angle simulation experiment results show that OCPA can effectively reduce the cost and delay compared with the benchmarks. At the same time, the achieved accuracy is close to the set long-term time-averaged value.

Due to the penetration of edge computing, a wide variety of workloads are sunk down to the network edge to alleviate huge pressure of the cloud. With the presence of high input workload dynamics and intensive edge resource contention, it is highly non-trivial for an edge proxy to optimize the scheduling of heterogeneous services with diverse QoS requirements. In general, online services should be quickly completed in a quite stable running environment to meet their tight latency constraint, while offline services can be processed in a loose manner for their elastic soft deadlines. To well coordinate such services at the resource-limited edge cluster, in this paper, we study an edge-centric resource provisioning optimization for dynamic online and offline services co-location, where the proxy seeks to maximize timely online service performances while maintaining satisfactory long-term offline service performances. However, intricate hybrid couplings for provisioning decisions arise due to heterogeneous constraints of the co-located services and their different time-scale performances. We hence first propose a reactive provisioning approach without requiring a prior knowledge of future system dynamics, which leverages a Lagrange relaxation for devising constraint-aware stochastic subgradient algorithm to deal with the challenge of hybrid couplings. To further boost the performance by integrating the powerful machine learning techniques, we also advocate a predictive provisioning approach, where the future request arrivals can be estimated accurately. With rigorous theoretical analysis and extensive trace-driven evaluations, we show the superior performance of our proposed algorithms for online and offline services co-location at the edge.

Privacy-preserving vector mean estimation is a crucial primitive in federated analytics. Existing practices usually resort to Local Differentiated Privacy (LDP) mechanisms that inject random noise into users' vectors when communicating with users and the central server. Due to the privacy-utility trade-off, the privacy budget has been widely recognized as the bottleneck resource that requires well provisioning. In this paper, we explore the possibility of privacy budget recycling and propose a novel Chained-DP framework enabling users to carry out data aggregation sequentially to recycle the privacy budget. We establish a sequential game to model the user interactions in our framework. We theoretically show the mathematical nature of the sequential game, solve its Nash Equilibrium, and design an incentive mechanism with provable economic properties. Our numerical simulation validates the effectiveness of Chained-DP, showing that it can significantly save privacy budget as well as lower estimation error compared to the traditional LDP mechanism.

Federated Learning (FL) is a distributed learning paradigm that can coordinate heterogeneous edge devices to perform model training without sharing private raw data. Many prior works have analyzed the FL convergence with respect to important hyperparameters, including batch size and aggregation frequency. However, adjusting the batch size and the number of local updates can affect the model performance, training time, and the cost of consuming computation and communication resources, in different and perhaps complex forms. Their joint effects have been overlooked and should be exploited to achieve accurate models with controllable operational expenditure. This paper proposes novel analytical models and optimization algorithms that leverage the interplay of batch size and aggregation frequency to navigate the trade-offs among convergence, cost, and completion time for FL. We first obtain a new convergence bound of the training error under heterogeneous training datasets across devices. Based on this bound, we derive closed-form solutions of a co-optimized batch size and aggregation frequency, a single configuration for all the devices. We then design an efficient exact algorithm for assigning different batch configurations across devices that can further improve the model accuracy to address the heterogeneity of both data and system characteristics. Further, we propose an adaptive control algorithm to dynamically adjust the solutions with estimated network states. Extensive experiments demonstrate the superiority of our offline optimal solutions and online adaptive algorithm.

With the increasing penetration of renewable energy and electric vehicles (EVs), the behavior of EVs' charging and discharging has shown great impact on the Micro Grid power load, motivating the development of Vehicle-to-Grid (V2G) technologies. However, the V2G market is still in its infancy, due to insufficient understanding of EV users' willingness and concerns. While many studies consider direct EV control, it's more realistic to indirectly affect users' behavior through monetary incentives. For better implementation flexibility, we advocate to display at charging piles strategically chosen incentives that are combined with electricity prices. Technically, this is the first model-free learning algorithm that can optimize incentives under unknown EV user reactions, increase the load control effectiveness and users' quality-of-service (QoS) simultaneously under a long-term incentive budget, and provide theoretical performance guarantees. We first construct a bi-level optimization framework to model the time-dependencies across our solutions. We then integrate primal-dual theories and upper-confidence bounds into reinforcement learning to balance power control and incentive consumption. A dynamic programming based algorithm is also proposed to maximize the aggregate user QoS. Finally, we prove bounded sub-optimality of our learning algorithm through theoretical analysis and conduct trace-driven simulations to demonstrate the advantages of our bi-level framework.

Deep learning has achieved tremendous success in various fields, but its significant computational demands make inference on mobile devices extremely challenging. To address this issue, we propose the COUPLE system, which enables heterogeneous processors to collaborate on mobile devices for accelerating video analytics. Additionally, we design the Co-Optimize strategy which utilizes the inference results of GPU to mitigate the accuracy loss caused by DSP. Experimental results demonstrate that COUPLE can improve the inference Average Precision by up to 5% compared to existing solutions.

Federated learning (FL) has recently emerged as a promising learning paradigm, that enables local gradient update and global model aggregation and synchronization. With the advantages of data privacy protection and communication overhead reduction, FL has been widely adopted in a multitude of edge intelligence and IoT applications. However, except the benefits, FL also suffers from stragglers effect and non-IID data, leading to unpredictable training delay and unstable convergence. Meanwhile, coding theory-based approaches have been proposed for stragglers mitigation in distributed computing, i.e., coded computing. Motivated by coded computing, there are several works introduce coding techniques into federated learning. In this paper, to further exploit the potential of coded federated learning, we propose HarFL, a stragglers resilient federated learning framework for non-IID data based on harmonic coding, which is suitable for general machine learning model with multivariate polynomial gradients and achieves a better stragglers mitigation than former coded federated learning schemes with the same coded computation redundancy. We first describe the basic harmonic coded federated learning framework with two phases: encoded data sharing and gradient results decoding. Moreover, we formulate an optimization problem aiming to maximize the successful probability of decoding, which is proved to be NP-hard. An efficient approximate algorithm with theoretical performance guarantee is also developed to mathematically solve the formulated problem. Finally, simulation results demonstrate the superiority of HarFL over several compared schemes.

Deep reinforcement learning (DRL) has been a successful paradigm in machine learning that enables solving complex control problems at the human level. However, the sampling and training efficiency of state-of-the-art DRL frameworks can not satisfy the stringent latency and throughput requirements of today’s mobile environments. Existing distributed and offline reinforcement learning algorithms along with the libraries for training acceleration are inherently designed for DRL tasks performed in the cloud rather than on distributed mobile devices, on which the computing resources are highly constrained, heterogeneous, and possibly dynamically changing. With the rise of edge computing and intelligence services, this paper presents RLink, a novel distributed training library to accelerate on-device deep reinforcement learning with inference knowledge at the edge. We leverage knowledge distillation to realize lightweight interaction between our on-device training task and the remote models that can provide inference knowledge. In this way, RLink is designed to be event-driven and agnostic to heterogeneous deep reinforcement learning algorithms and libraries. To tackle the communication bottleneck, a novel asynchronous sampling algorithm is proposed to facilitate real-time training in RLink. Tuned for unstable-connected mobile devices, RLink is robust and efficient by using a semantic-aware communication pipeline for lossless data compression. Extensive experimental results show that, compared with state-of-the-art algorithms and libraries, RLink can accelerate deep reinforcement learning at the edge with up to decuple speedups in convergence and ideal computational performance.

Deep learning (DL) powered real-time applications usually need continuous training using data streams generated geographically. Enabling data offloading among computation nodes through model training is promising to mitigate the problem that devices generating large datasets may have low computation capability. However, offloading can compromise model convergence and incur communication costs, which must be balanced with the cost spent on computation and model synchronization. Therefore, this paper proposes EdgeC3, a novel framework that can optimize the frequency of model aggregation and dynamic offloading for continuously generated data streams, navigating the trade-off between long-term accuracy and cost. We first provide a new error bound to capture the impacts of data dynamics that are varying over time and heterogeneous across devices. Based on the bound, we design a two-timescale online optimization framework. We periodically learn the synchronization frequency to adapt with uncertain future offloading and network changes. In the finer timescale, we manage online offloading by extending Lyapunov optimization techniques to handle an unconventional setting, where our long-term global constraint can have abruptly changed aggregation frequencies that are decided in the longer timescale. Finally, we theoretically prove the convergence of EdgeC3 by integrating the coupled effects of our two-timescale decisions, and we demonstrate its advantage through extensive experiments.

As the killer application of edge computing, video analytics typically involves multiple vision components in the pipeline, which together determine the quality of service (QoS) for users. By exploiting diverse resource demands of different components, a fine-grained cross-layer orchestration with QoS-aware configuration adaptation can further boost the system efficiency of heterogeneous resources. Thus, we study a video analytics pipeline system with vertical and horizontal resource collaboration across device-edge-cloud hierarchy to achieve QoS-aware cost optimization. To judiciously match the component diversity and the resource heterogeneity, we explore smooth configuration adaptation to model a mixed-integer nonlinear problem, which jointly optimizes long-term resource cost and QoS (including accuracy and latency). However, it is nontrivial to efficiently solve such a NP-hard problem in an online manner without the future information as a prior knowledge due to the time-coupling deployment cost caused by fluctuating input traffic. To address the above challenges, we decouple the intractable problem according to the traffic routing constraints. By leveraging the lazy-switching method, we derive the component orchestration decisions for the decoupled subproblems in each slot and further design a dependent rounding scheme to obtain an efficient feasible solution while guaranteeing the knapsack resource constraints. We rigorously analyze the performance guarantee of our online algorithms by a parameterized competitive ratio, and further verify the empirical performance of our approach through extensive trace-driven experiments.

Edge computing can reduce response time and communication pressure by caching data in edge server which is closed to the user. The development of 5G applications has posed strict requirements on various performance metrics of communication, which brings more and more challenges to the design of edge networks. In this paper, we focus on a multi-hop cache update system with multiple caches in series, which models a real-world system consisting of cloud, macro base stations, small base stations, and users. The base stations cache data and respond to users’ request. We derived closed-form expressions of average information freshness and average transmission delay for fetching data from the given level of cache and jointly optimize them. Then, we propose an algorithm based on alternating maximization to solve the optimization problem. The experimental results show that the proposed method can always achieve the optimum under different parameter settings. In addition, we also analyze the variation of latency and information freshness under the influence of weight parameters, which illustrates the importance of joint optimization of latency and information freshness.

Deploying federated learning (FL) in wireless networks faces the critical challenge of communication bottlenecks. To address this issue, in this paper, we consider an over-the-air computation (AirComp) assisted hierarchical FL (HFL) framework, where a cloud-edge-device-based three-tier network architecture is constructed to train a global model. We first theoretically characterize the convergence of the AirComp-assisted HFL framework and formulate a combinatorial optimization problem that jointly optimizes the edge interval control and local device transceiver design to minimize the convergence upper bound to boost the overall learning performance and reduce communication cost. We show that the formulated optimization problem can be decoupled into an edge interval control problem and a transceiver design problem, which can be tackled by developing a relaxation and rounding algorithm and an alternating Lyapunov drift-based algorithm, respectively. Extensive simulations demonstrate that our proposed algorithm significantly outperforms the baseline schemes.

Edge AI is an emerging paradigm that leverages edge computing to pave the last mile delivery of artificial intelligence. To satisfy the stringent timeliness and energy-efficiency requirements of emerging edge AI tasks, specialized AI accelerator of Neural Processing Units (NPU) have been widely equipped by edge nodes. Compared to the traditional centralized processing units (CPU), NPU has better performance and energy-efficiency. However, these benefits come at the cost of reduced inference accuracy. As a result, existing coarse-grained scheduling mechanisms that schedule a whole DNN task to either the CPU or NPU are unable to make the best use of NPU. To address this issue, we propose an online NPU-CPU co-inference scheduling mechanism to schedule the DNN task at the fine-grained layer level, and thus to fully utilize the performance, accuracy, and power diversities of the NPU and CPU. By applying Lyapunov optimization to schedule the network layers dynamically, our proposed online scheduling mechanism is able to ensure the real-time inference speed and cap the long-term time-averaged power consumption, while still approximately minimizes the long-term inference accuracy loss. Via rigorous theoretical analysis as well as realistic trace-driven simulations, we demonstrate the effectiveness of our proposed online scheduling mechanism.

While 5G networks have accelerated the development of the Internet of Things (IoT), they have also introduced a large number of vulnerable IoT devices into the network, which would lead to severe Distributed Denial-of-Service (DDoS) attacks. The newly emerging DDoS attack methods generally have a shorter duration, which imposes higher requirements for the response time of DDoS mitigation technologies. Existing DDoS defense methods cannot achieve real-time detection due to the difficulty of reducing the delay of feature extraction and large-scale data processing. In this article, we focus on the timeliness of DDoS detection and mitigation. We hope that deploying effective defense countermeasures at the source side will block the majority of DDoS attack traffic in real time before it enters the data network (DN). To this end, we propose a real-time DDoS defense framework based on multidomain collaboration that combines multisource information to detect attack sessions with high accuracy in 5G networks. To operate the framework at line rate, we propose an optimal packet sampling strategy based on the accurate session size estimation, which can greatly reduce the detection overhead while ensuring good accuracy. In a typical scenario with an attack session size larger than 10, this method can achieve a 99% detection rate while reducing the packet inspection rate (PIR) to less than 37%.

Due to the heterogeneity of Internet of Things (IoT) devices and the diversity of IoT communication protocols, it is challenging to model the communication behaviors of IoT devices to facilitate attack defense. Considering the complex correlation between the IoT device types and the patterns of their communication behaviors, one possible solution is to cluster IoT devices into different types based on the characteristics of their communication behaviors and deal with each type, respectively. However, IoT traffic includes a significant proportion of abnormal traffic, such as attack traffic sourcing from compromised devices, which cannot reflect the behavioral characteristics of the source device. In this article, we propose a Transformer-based IoT device-type identification method to address the above challenges. Specifically, our approach consists of three main components. First, we classify the traffic data from IoT devices into normal and abnormal types by a Transformer-based traffic diagnosis model. Next, another Transformer-based model is adopted on the normal traffic to identify the IoT device type. Finally, considering the immutability of IoT device types, a results-ensemble algorithm is designed to improve the accuracy of IoT device-type identification. Experimental results verify the effectiveness of our method, which brings a noticeable improvement in terms of both accuracy and macro$F1$-score compared to other methods. Moreover, by applying the results-ensemble algorithm in the test phase, we can achieve 100% accuracy under certain conditions.

Mobile-edge computing (MEC) is widely envisioned as a promising technique for provisioning artificial intelligence (AI) capability for resource-limited Internet of Things (IoT) devices by leveraging edge servers (ESs) for executing deep neural network (DNN) inference tasks in proximity. However, scheduling DNN inference tasks at the network edge under unknown system dynamics (e.g., uncertain availability of ESs) may suffer from failures, making it difficult to guarantee reliable services for the IoT device. To overcome this challenge, we propose a reliability-aware online scheduling scheme for DNN inference tasks in MEC by leveraging both online feedback and offline data to learn the uncertain availability of ESs to maximize both the inference accuracy and service reliability of DNN inference tasks (i.e., the number of DNN inference tasks processed during the system span). We first formulate the reliability-aware DNN inference tasks scheduling problem as a novel constrained combinatorial multiarmed bandit (CMAB) problem. Then by integrating the Lyapunov optimization technique, bandit learning, approximated submodular maximization, and historical data organically, we design a reliability-aware task scheduling scheme with a bandit learning (RTBL) algorithm to solve this problem. Unfortunately, even with an accurate prediction of the system uncertainties, the task scheduling problem is still NP-hard. To deal with it, we, therefore, design an advanced approximation algorithm based on the submodularity of the scheduling problem which obtains a near-optimal solution and provides a satisfactory performance guarantee. Finally, we conduct rigorous theoretical analysis and race-driven simulations to show RTBL’s brilliant performance.

Provisioning dynamic machine learning (ML) inference as a service for artificial intelligence (AI) applications of edge devices faces many challenges, including the trade-off among accuracy loss, carbon emission, and unknown future costs. Besides, many governments are launching carbon emission rights (CER) for operators to reduce carbon emissions further to reverse climate change. Facing these challenges, to achieve carbon-aware ML task offloading under limited carbon emission rights thus to achieve green edge AI, we establish a joint ML task offloading and CER purchasing problem, intending to minimize the accuracy loss under the long-term time-averaged cost budget of purchasing the required CER. However, considering the uncertainty of the resource prices, the CER purchasing prices, the carbon intensity of sites, and ML tasks’ arrivals, it is hard to decide the optimal policy online over a long-running period time. To overcome this difficulty, we leverage the two-timescale Lyapunov optimization technique, of which the T-slot drift-plus-penalty methodology inspires us to propose an online algorithm that purchases CER in multiple timescales (on-preserved in carbon future market and on-demanded in the carbon spot market) and makes decisions about where to offload ML tasks. Considering the NP-hardness of the T-slot problems, we further propose the resource-restricted randomized dependent rounding algorithm to help to gain the near-optimal solution with no help of any future information. Our theoretical analysis and extensive simulation results driven by the real carbon intensity trace show the superior performance of the proposed algorithms.

Edge intelligence has arisen as a promising computing paradigm for supporting miscellaneous smart applications that rely on machine learning techniques. While the community has extensively investigated multi-tier edge deployment for traditional deep learning models (e.g. CNNs, RNNs), the emerging Graph Neural Networks (GNNs) are still under exploration, presenting a stark disparity to its broad edge adoptions such as traffic flow forecasting and location-based social recommendation. To bridge this gap, this paper formally studies the cost optimization for distributed GNN processing over a multi-tier heterogeneous edge network. We build a comprehensive modeling framework that can capture a variety of different cost factors, based on which we formulate a cost-efficient graph layout optimization problem that is proved to be NP-hard. Instead of trivially applying traditional data placement wisdom, we theoretically reveal the structural property of quadratic submodularity implicated in GNN’s unique computing pattern, which motivates our design of an efficient iterative solution exploiting graph cuts. Rigorous analysis shows that it provides parameterized constant approximation ratio, guaranteed convergence, and exact feasibility. To tackle potential graph topological evolution in GNN processing, we further devise an incremental update strategy and an adaptive scheduling algorithm for lightweight dynamic layout optimization. Evaluations with real-world datasets and various GNN benchmarks demonstrate that our approach achieves superior performance over de facto baselines with more than 95.8% cost reduction in a fast convergence speed.

Federated learning (FL) is a promising distributed framework for collaborative artificial intelligence model training while protecting user privacy. A bootstrapping component that has attracted significant research attention is the design of incentive mechanism to stimulate user collaboration in FL. The majority of works adopt a broker-centric approach to help the central operator to attract participants and further obtain a well-trained model. Few works consider forging participant-centric collaboration among participants to pursue an FL model for their common interests, which induces dramatic differences in incentive mechanism design from the broker-centric FL. To coordinate the selfish and heterogeneous participants, we propose a novel analytic framework for incentivizing effective and efficient collaborations for participant-centric FL. Specifically, we respectively propose two novel game models for contribution-oblivious FL (COFL) and contribution-aware FL (CAFL), where the latter one implements a minimum contribution threshold mechanism. We further analyze the uniqueness and existence for Nash equilibrium of both COFL and CAFL games and design efficient algorithms to achieve equilibrium solutions. Extensive performance evaluations show that there exists free-riding phenomenon in COFL, which can be greatly alleviated through the adoption of CAFL model with the optimized minimum threshold.

Mobile Edge Computing (MEC), envisioned as a cloud extension, pushes cloud resource from the network core to the network edge, thereby meeting the stringent service requirements of many emerging computation-intensive mobile applications. Many existing works have focused on studying the system-wide MEC service placement issues, personalized service performance optimization yet receives much less attention. As motivated, in this paper we propose a novel adaptive user-managed service placement mechanism, which jointly optimizes a users perceived-latency and service migration cost, weighted by user-specific preferences. We first formulate the user-managed dynamic service placement process with limited system information as a contextual multi-armed bandit learning problem. In particular, we investigate both cases without and with neighboring edge feedbacks, where the later considers edge information sharing for more informed decision making. For both cases, we design lightweight Thompson-sampling based online learning algorithms, which can efficiently assist the user to make adaptive service placement decisions. We further conduct a novel information-directed theoretical analysis on the regret bound of the proposed online learning algorithms and reveal the structural impact of edge information sharing. Extensive evaluations demonstrate the superior performance gain of the proposed adaptive user-managed service placement mechanism over existing learning schemes.

Network Function Virtualization (NFV) provides the possibility to implement complex network functions from dedicated hardware to software instances called Virtual Network Functions (VNF) by leveraging the virtualization technology. Service Function Chaining (SFC) is therefore defined as a chain-ordered set of placed VNFs that handles the traffic of the delivery and control of a specific application. Due to the advantages of flexibility, efficiency, scalability, and short deployment cycles, NFV has been widely recognized as the next-generation network service provisioning paradigm. In this paper, we study the problem of online SFC control across geo-distributed datacenters, which is to dynamically place required VNFs on datacenter nodes and find routing paths between each adjacent VNF pair for each NFV service flow that varies over time. To that end, we first formulate this problem as an offline optimization problem whose goal is to minimize the average delay such that each datacenter's average cost does not exceed a given expense value. Considering that the offline optimization requires complete offline network information which is difficult to obtain or predict in practice, we present an online SFC control framework without requiring any future information about the traffic demands. More specifically, we leverage the Lyapunov optimization technique to formulate the problem as a series of one-time slot offline optimization problems and then apply a primal-decomposition method to solve each one-time slot problem. Simulation results reveal that our proposed online SFC control framework can efficiently reduce long-term average delay while keeping datacenter's long-term average cost consumption low.

The advancement of smartphone sensors and wearable devices has enabled a new paradigm for smart human activity recognition (HAR), which has a broad range of applications in healthcare and smart cities. However, there are four challenges,privacy preservation,label scarcity,real-timing, andheterogeneity patterns, to be addressed before HAR can be more applicable in real-world scenarios. To this end, in this paper, we propose a personalized federated HAR framework, namedFedHAR, to overcome all the above obstacles. Specially, as federated learning,FedHARperforms distributed learning, which allows training data to be kept local to protect users’ privacy. Also, for each client without activity labels, inFedHAR, we design an algorithm to compute unsupervised gradients under theconsistency trainingproposition and an unsupervised gradient aggregation strategy is developed for overcoming the concept drift and convergence instability issues in online federated learning process. Finally, extensive experiments are conducted using two diverse real-world HAR datasets to show the advantages ofFedHARover state-of-the-art methods. In addition, when fine-tuning each unlabeled client, personalizedFedHARcan achieve additional 10% improvement across all metrics on average.

Cross-silo federated learning (FL) is a distributed learning approach where clients of the same interest train a global model cooperatively while keeping their local data private. The success of a cross-silo FL process requires active participation of many clients. Different from cross-device FL, clients in cross-silo FL are usually organizations or companies which may execute multiple cross-silo FL processes repeatedly due to their time-varying local data sets, and aim to optimize their long-term benefits by selfishly choosing their participation levels. While there has been some work on incentivizing clients to join FL, the analysis of clients’ long-term selfish participation behaviors in cross-silo FL remains largely unexplored. In this paper, we analyze the selfish participation behaviors of heterogeneous clients in cross-silo FL. Specifically, we model clients’ long-term selfish participation behaviors as an infinitely repeated game, with the stage game being a selfish participation game in one cross-silo FL process (SPFL). For the stage game SPFL, we derive the unique Nash equilibrium (NE), and propose a distributed algorithm for each client to calculate its equilibrium participation strategy. We show that at the NE, clients fall into at most three categories: (i)free riderswho do not perform local model training, (ii) a uniquepartial contributor(if exists) who performs model training with part of its local data, and (iii)contributorswho perform model training with all their local data. The existence of free riders has a detrimental effect on achieving a good global model and sustaining other clients’ long-term participation. For the long-term interactions among clients, we derive a cooperative strategy for clients which minimizes the number of free riders while increasing the amount of local data for model training. We show that enforced by a punishment strategy, such a cooperative strategy is a subgame perfect Nash equilibrium (SPNE) of the infinitely repeated game, under which some clients who are free riders at the NE of the stage game choose to be (partial) contributors. We further propose an algorithm to calculate the optimal SPNE which minimizes the number of free riders while maximizing the amount of local data for model training. Simulation results show that our derived optimal SPNE can effectively reduce the number of free riders by up to$99.3\%$and increase the amount of local data for model training by up to$82.3\%$.

Crowdsourcing has found a wide variety of applications, including spectrum sensing, traffic monitoring, as well as data annotation for machine learning based data analytics. To improve data accuracy and cost-effectiveness, workers’ data quality can be learned from their data in an online manner, which can be used for task assignment and data aggregation. However, crowdsourcing is vulnerable to data poisoning attacks, where the attacker reports malicious data to reduce aggregated data accuracy. In this paper, we study malicious data attacks on dynamic crowdsourcing where tasks are assigned and performed sequentially, and we explore online quality learning as a defense mechanism against the attack by finding malicious workers with low quality. We first focus on the asymptotic setting where workers’ quality is accurately learned by the requester, based on which we then turn to the general non-asymptotic setting where the quality is estimated online with errors. For each setting, we first characterize the conditions under which the attack strategy can effectively reduce the aggregated data accuracy. Our results show that the malicious noise variance needs to be within a certain range for the attack to be effective. Then we analyze the harm of effective attack strategies. It reveals that the regret of the online quality learning algorithm can be substantially increased from$\mathcal {O}(\log ^2T)$(upper bound) to$\Omega (T)$(lower bound) due to effective attacks. To further mitigate the attack, we also study median and maximum influence of estimation based data aggregation as defense mechanisms. Our results provide useful insights on the impacts of data poisoning attacks when online quality learning is used to defend against the attack. We evaluate the proposed attacks and defenses via extensive simulation results based on real-world data, which demonstrate the effectiveness of the attacks and defenses.

The accelerating convergence of artificial intelligence and edge computing has sparked a recent wave of interest in edge intelligence. While pilot efforts focused on edge DNN inference serving for a single user or DNN application, scaling edge DNN inference serving to multiple users and applications is however nontrivial. In this paper, we propose an online optimization framework EdgeAdaptor for multi-user and multi-application edge DNN inference serving at scale, which aims to navigate the three-way trade-off between inference accuracy, latency, and resource cost via jointly optimizing the application configuration adaption, DNN model selection and edge resource provisioning on-the-fly. The underlying long-term optimization problem is difficult since it is NP-hard and involves future uncertain information. To address these dual challenges, we fuse the power of online optimization and approximate optimization into a joint optimization framework, via i) decomposing the long-term problem into a series of single-shot fractional problems with a regularization technique, and ii) rounding the fractional solution to a near-optimal integral solution with a randomized dependent scheme. Rigorous theoretical analysis derives a parameterized competition ratio of our online algorithms, and extensive trace-driven simulations verify that its empirical value is no larger than 1.4 in typical scenarios.

Data reporters have privacy concerns when they are requested to contribute personal data to a data collector. Such privacy concerns are strengthened by data correlation and social relationship, as the data correlation could inevitably cause privacy issues to their socially-connected individuals who even do not report the data. However, both factors are hard to quantify precisely in practice due to their private nature. Such an incomplete information situation poses great challenges for the data reporters to determine their coupled privacy-preserving strategies and for the data collector to choose a proper privacy-preserving mechanism. This motivates us to propose a novel Bayesian game-theoretic framework to analyze the data reporters’ behaviors. We show that the game has a symmetric Bayesian Nash Equilibrium (BNE) with a threshold structure, which builds a connection between the data reporter’s action and privacy concern under incomplete information. The complicated relationship between the BNE and the data collector’s strategy makes it difficult to solve the data collector’s optimization problem. However, by exploiting the unimodal feature of the problem, we present a low-complexity algorithm to compute the optimal privacy-preserving mechanism. Through analytical and numerical studies, we find that the lack of complete information could cause the data reporters to adopt more conservative strategies but make the data collector adopt a less conservative mechanism, resulting in an overall privacy protection degradation. The simulations further demonstrate that the degradation could be alleviated by stronger data correlation and social relationship, and a higher probability of serious privacy concerns.

In vanilla Federated Learning (FL) systems, a centralized parameter server (PS) is responsible for collecting, aggregating and distributing model parameters with decentralized clients. However, the communication link of a single PS can be easily overloaded by concurrent communications with a massive number of clients. To overcome this drawback, multiple PSes can be deployed to form a parallel FL (PFL) system, in which each PS only communicates with a subset of clients and its neighbor PSes. On one hand, each PS conducts iterations with clients in its subset. On the other hand, PSes communicate with each other periodically to mix their parameters so that they can finally reach a consensus. In this paper, we propose a novel parallel federated learning algorithm called Fed-PMA, which optimizes such parallel FL under constrained communications by conducting parallel parameter mixing and averaging with theoretic guarantees. We formally analyze the convergence rate of Fed-PMA with convex loss, and further derive the optimal number of times each PS should mix with its neighbor PSes so as to maximize the final model accuracy within a fixed span of training time. Theoretical study manifests that PSes should mix their parameters more frequently if the connection between PSes is sparse or the time cost of mixing is low. Inspired by our analysis, we propose the Fed-APMA algorithm that can adaptively determine the near-optimal number of mixing times with non-convex loss under dynamic communication conditions. Extensive experiments with realistic datasets are carried out to demonstrate that both Fed-PMA and its adaptive version Fed-APMA significantly outperform the state-of-the-art baselines.

Machine learning (ML) tasks are one of the major workloads in today's edge computing networks. Existing edge-cloud schedulers allocate the requested amounts of resources to each task, falling short of best utilizing the limited edge resources for ML tasks. This paper proposesTapFinger, a distributed scheduler for edge clusters that minimizes the total completion time of ML tasks through co-optimizing task placement and fine-grained multi-resource allocation. To learn the tasks’ uncertain resource sensitivity and enable distributed scheduling, we adopt multi-agent reinforcement learning (MARL) and propose several techniques to make it efficient, including a heterogeneous graph attention network as the MARL backbone, a tailored task selection phase in the actor network, and the integration of Bayes’ theorem and masking schemes. We first implement asingle-task schedulingversion, which schedules at most one task each time. Then we generalize to themulti-task schedulingcase, in which a sequence of tasks is scheduled simultaneously. Our design can mitigate the expanded decision space and yield fast convergence to optimal scheduling solutions. Extensive experiments using synthetic and test-bed ML task traces show thatTapFingercan achieve up to 54.9% reduction in the average task completion time and improve resource efficiency as compared to state-of-the-art schedulers.

Federated learning (FL) is a promising paradigm that enables collaboratively learning a shared model across massive clients while keeping the training data locally. However, for many existing FL systems, clients need to frequently exchange model parameters of large data size with the remote cloud server directly via wide-area networks (WAN), leading to significant communication overhead and long transmission time. To mitigate the communication bottleneck, we resort to the hierarchical federated learning paradigm of HiFL, which reaps the benefits of mobile edge computing and combines synchronous client-edge model aggregation and asynchronous edge-cloud model aggregation together to greatly reduce the traffic volumes of WAN transmissions. Specifically, we first analyze the convergence bound of HiFL theoretically and identify the key controllable factors for model performance improvement. We then advocate an enhanced design of HiFlash by innovatively integrating deep reinforcement learning based adaptive staleness control and heterogeneity-aware client-edge association strategy to boost the system efficiency and mitigate the staleness effect without compromising model accuracy. Extensive experiments corroborate the superior performance of HiFlash in model accuracy, communication reduction, and system efficiency.

The space-air-ground integrated network (SAGIN), one of the key technologies for next-generation mobile communication systems, can facilitate data transmission for users all over the world, especially in some remote areas where vast amounts of informative data are collected by Internet of remote things (IoRT) devices to support various data-driven artificial intelligence (AI) services. However, training AI models centrally with the assistance of SAGIN faces the challenges of highly constrained network topology, inefficient data transmission, and privacy issues. To tackle these challenges, we first propose a novel topology-aware federated learning framework for the SAGIN, namely Olive Branch Learning (OBL). Specifically, the IoRT devices in the ground layer leverage their private data to perform model training locally, while the air nodes in the air layer and the ring-structured low earth orbit (LEO) satellite constellation in the space layer are in charge of model aggregation (synchronization) at different scales. To further enhance communication efficiency and inference performance of OBL, an efficient Communication and Non-IID-aware Air node-Satellite Assignment (CNASA) algorithm is designed by taking the data class distribution of the air nodes as well as their geographic locations into account. Furthermore, we extend our OBL framework and CNASA algorithm to adapt to more complex multi-orbit satellite networks. We analyze the convergence of our OBL framework and conclude that the CNASA algorithm contributes to the fast convergence of the global model. Extensive experiments based on realistic datasets corroborate the superior performance of our algorithm over the benchmark policies.

Federated learning (FL) leverages the private data and computing power of multiple clients to collaboratively train a global model. Many existing FL algorithms over wireless networks adopting synchronous model aggregation suffer from the straggler issue, due to the heterogeneity of local computing power and channel conditions. To address this issue, we in this paper advocate an asynchronous FL framework with adaptive client selection for training latency minimization, taking into account the client availability and long-term fairness. We consider a practical scenario, where the channel conditions and the locally available computing power are not known in prior. This makes the client selection problem challenging, as the training latency consists of the uplink/downlink transmission time and the local training time. To this end, we tackle the asynchronous client selection problem in an online manner by converting the latency minimization problem into a multi-armed bandit problem, and leverage the upper confidence bound policy and virtual queue technique in Lyapunov optimization to solve the problem. We theoretically show that the proposed algorithm achieves sub-linear regret performance, ensures long-term fairness, and guarantees training convergence. Results show that the proposed algorithm can reduce the training time by up to 50% when compared to the baseline algorithms.

In this paper, we consider communication-efficient over-the-air federated learning (FL), where multiple edge devices with non-independent and identically distributed datasets perform multiple local iterations in each communication round and then concurrently transmit their updated gradients to an edge server over the same radio channel for global model aggregation using over-the-air computation (AirComp). We derive the upper bound of the time-average norm of the gradients to characterize the convergence of AirComp-assisted FL, which reveals the impact of the model aggregation errors accumulated over all communication rounds on convergence. Based on the convergence analysis, we formulate an optimization problem to minimize the upper bound to enhance the learning performance, followed by proposing an alternating optimization algorithm to facilitate the transceiver design for AirComp-assisted FL. As the alternating optimization algorithm suffers from high computation complexity, we further develop a knowledge-guided learning algorithm that exploits the structure of the analytic expression of the transmit power to achieve computation-efficient transceiver design. Simulation results demonstrate that the proposed knowledge-guided learning algorithm achieves a comparable performance as the alternating optimization algorithm, but with a much lower computation complexity. Moreover, both proposed algorithms outperform the baseline methods in terms of convergence speed and test accuracy.

Due to the heterogeneity of Internet of Things (IoT) devices and the diversity of IoT communication protocols, it is challenging to defend against malicious traffic from IoT devices. In this paper, a novel malicious traffic detection method is proposed based on the deep learning method. Specifically, a Transformer-based encoder is designed to automatically select key features of IoT traffic for the detection task, which avoids the cumbersome feature screening process that has been widely used in traditional machine learning methods. To address the complexity of the feature space and improve the efficiency of model training, we exploit the correlation between the characteristics of malicious traffic and the device type of IoT bots to further improve the detection accuracy by introducing a device classification auxiliary loss in the training phase. Experimental results show that our method outperforms the state-of-the-art machine learning-based methods in terms of accuracy, precision, recall and f1-score on real IoT traffic traces. In addition, the benefit of device type information on detection efficiency is verified.

Accurate navigation is of paramount importance to ensure flight safety and efficiency for autonomous drones. Recent research starts to use Deep Neural Networks (DNN) to enhance drone navigation given their remarkable predictive capability for visual perception. However, existing solutions either run DNN inference tasks on drones in-situ, impeded by the limited onboard resource, or offload the computation to external servers which may incur large network latency. Few works consider jointly optimizing the offloading decisions along with image transmission configurations and adapting them on the fly. In this paper, we propose AdaDrone, an edge computing assisted drone navigation framework that can dynamically adjust task execution location, input resolution, and image compression ratio in order to achieve low inference latency, high prediction accuracy, and long flight distances. Specifically, we first augment state-of-the-art convolutional neural networks for drone navigation and define a novel metric called Quality of Navigation as our optimization objective which can effectively capture the above goals. We then design a deep reinforcement learning (DRL) based neural scheduler for which an information encoder is devised to reshape the state features and thus improve its learning ability. We finally implement a prototype of our framework wherein a drone board for navigation and scheduling control interacts with edge servers for task offloading and a simulator for performance evaluation. Extensive experimental results show that AdaDrone can reduce end-to-end latency by 28.06% and extend the flight distance by up to 27.28% compared with non-adaptive solutions.

Federated Learning (FL) has been a promising paradigm in distributed machine learning that enables in-situ model training and global model aggregation. While it can well preserve private data for end users, to apply it efficiently on IoT devices yet suffer from their inherent variants: their available computing resources are typically constrained, heterogeneous, and changing dynamically. Existing works deploy FL on IoT devices by pruning a sparse model or adopting a tiny counterpart, which alleviates the workload but may have negative impacts on model accuracy. To address these issues, we propose Eco-FL, a novel Edge Collaborative pipeline based Federated Learning framework. On the client side, each IoT device collaborates with trusted available devices in proximity to perform pipeline training, enabling local training acceleration with efficient augmented resource orchestration. On the server side, Eco-FL adopts a novel grouping-based hierarchical architecture that combines synchronous intra-group aggregation and asynchronous inter-group aggregation, where a heterogeneity-aware dynamic grouping strategy that jointly considers response latency and data distribution is developed. To tackle the resource fluctuation during the runtime, Eco-FL further applies an adaptive scheduling policy to judiciously adjust workload allocation and client grouping at different levels. Extensive experimental results using both prototype and simulation show that, compared to state-of-the-art methods, Eco-FL can upgrade the training accuracy by up to 26.3%, reduce the local training time by up to 61.5%, and improve the local training throughput by up to 2.6 ×.

Existing machine learning (ML) model marketplaces generally require data owners to share their raw data, leading to serious privacy concerns. Federated learning (FL) can partially alleviate this issue by enabling model training without raw data exchange. However, data owners are still susceptible to privacy leakage from gradient exposure in FL, which discourages their participation. In this work, we advocate a novel differentially private FL (DPFL)-based ML model marketplace. We focus on the broker-centric design. Specifically, the broker first incentivizes data owners to participate in model training via DPFL by offering privacy protection as per their privacy budgets and explicitly accounting for their privacy costs. Then, it conducts optimal model versioning and pricing to sell the obtained model versions to model buyers. In particular, we focus on the broker’s profit maximization, which is challenging due to the significant difficulties in the revenue characterization of model trading and the cost estimation of DPFL model training. We propose a two-layer optimization framework to address it, i.e., revenue maximization and cost minimization under model quality constraints. The latter is still challenging due to its non-convexity and integer constraints. We hence propose efficient algorithms, and their performances are both theoretically guaranteed and empirically validated.

Recent advances in deep learning models have pushed Super-Resolution (SR) techniques to an unprecedented altitude, enabling high-quality image rendering with variable scaling size and natural fidelity. To deploy them on resource-constrained mobile devices, however, confronts significant chal-lenges of excessively long latency and poor user experience. To this end, we propose Apie, an edge-assisted adaptive image rendering system that allows low-latency, progressive image enhancement for a smooth user experience. Apie adopts a data parallel strategy across the end device and the edge server, along with a residual learning mechanism to judiciously retrieve information for SR models. Besides, a novel progressive image reconstruction is developed by exploiting content-aware image blocking and incremental image rendering, towards improved quality of user experience. Furthermore, Apie can dynamically adjust the choice of employed SR models with respect to the networking conditions, striking a good balance upon the latency-quality trade-off. Extensive evaluations show that Apie performs 7.33x faster than on-device GPU execution and 1.42x faster compared to the partial offloading method, while achieves 2.84dB higher PSNR compared to the interpolation method using conventional JPEG image compression and 0.74dB higher PSNR compared to the partial offloading method.

Graph Neural Networks (GNNs) have gained growing interest in miscellaneous applications owing to their outstanding ability in extracting latent representation on graph structures. To render GNN-based service for IoT-driven smart applications, the traditional model serving paradigm resorts to the cloud by fully uploading the geo-distributed input data to the remote datacenter. However, our empirical measurements reveal the significant communication overhead of such cloud-based serving and highlight the profound potential in applying the emerging fog computing. To maximize the architectural benefits brought by fog computing, in this paper, we present Fograph, a novel distributed real-time GNN inference framework that leverages diverse resources of multiple fog nodes in proximity to IoT data sources. By introducing heterogeneity-aware execution planning and GNN-specific compression techniques, Fograph tailors its design to well accommodate the unique characteristics of GNN serving in fog environment. Prototype-based evaluation and case study demonstrate that Fograph significantly outperforms the state-of-the-art cloud serving and vanilla fog deployment by up to 5.39 × execution speedup and 6.84 × throughput improvement.

The link flooding attack (LFA) has emerged as a new category of distributed denial of service (DDoS) attacks in recent years. Along with the massive deployment of low-cost insecure Internet-of-Things (IoT) devices, the fast proliferation of IoT botnets dramatically increases the risk of LFAs. However, how to efficiently defend against LFAs in IoT still remains as an open problem. To overcome this challenge, we model the interaction between an LFA attacker and the network manager as a two-person Bayesian game in this article to precisely characterize the behaviors of both sides. Then, the rational behaviors of the attacker and the optimal strategies of the defender are unveiled by deriving the Bayesian Nash equilibrium (BNE). Inspired by the obtained BNEs, a cost-effective decision framework is proposed for the defender to make defense decisions. Furthermore, we numerically analyze the effect of all the related factors and present feasible suggestions to deter attack motivations fundamentally. Experimental results demonstrate that the proposed method not only consistently outperforms baseline methods in terms of the defender’s utilities under different attack intensities, but also is robust to the changes in important parameters, including the value of benign traffic and the latency of traffic scrubbing.

The proliferation of Distributed Denial of Service (DDoS) attacks in Internet of Things (IoT) not only threatens the security of digital devices and infrastructure but also severely degrades IoT system performance due to the overly consumed network resources. With the knowledge of identity information of devices and signaling data, Internet service providers (ISPs) can detect and block DDoS traffic by monitoring the upstream IoT packets, and thereby, improve network efficiency. However, inspecting all data packets online for DDoS detection will significantly increase both the network delay and the computational overhead. Therefore, the packet sampling strategy is crucial for the defenders to detect DDoS attacks. To this end, this article formulates a Stackelberg game model to analyze the collaborative IoT packet sampling against DDoS attacks. Through the equilibrium analysis of the DDoS game, we derive the lower bound of packet sampling rate (PSR) that can effectively deter potential attackers. Unlike traditional offline detection, our proposed packet sampling strategy can support both the online detection and proactive prevention of DDoS traffic. As a use case, a multipoint DDoS defense framework is developed to address the IP spoofing in 5G networks based on the proposed packet sampling strategy, which deters DDoS attacks and reduces the packet sampling cost, and thereby, maximizes the IoT utility, compared with existing methods. In typical reflection attacks (in which no more than five packets of response are triggered by a request packet), our proposed scheme not only reduces more than 70% of the sampling rate but also demonstrates superior robustness against boundary condition variation.

With the rapid development of the Internet, more and more computing-intensive applications with high requirements on computing delay and energy consumption have emerged. Recently, the use of mobile-edge computing servers for auxiliary computing is considered as an effective way to reduce latency and energy consumption. In addition, applications such as autonomous driving will generate a large number of repetitive tasks. Using a cache to store the computational results of popular tasks can avoid the overhead caused by repetitive processing. In this article, we study the problem of computation offloading for users in multiple regions. The optimization goal is expressed as choosing offloading strategies and caching strategies to minimize the total delay and energy consumption of all regions. We first use the deep reinforcement learning (DRL) deep deterministic policy gradient (DDPG) framework to solve the problem of computational offloading in a single region. We also show the inefficiency of existing collaborative caching approaches in multiple regions, and propose a new collaborative caching algorithm (CCA) to improve the overall cache hit rate of the system. Finally, we integrate the DDPG and CCA algorithms to form a holistic efficient caching and offloading strategy for all regions. The simulation results show that the proposed algorithm can significantly improve the cache hit rate, and has an excellent performance in reducing the total system overhead.

Edge-computing-enhanced Internet of Vehicles (EC-IoV) enables ubiquitous data processing and content sharing among vehicles and terrestrial edge computing (TEC) infrastructures (e.g., 5G base stations and roadside units) with little or no human intervention, and plays a key role in the intelligent transportation systems. However, EC-IoV is heavily dependent on the connections and interactions between vehicles and TEC infrastructures, thus will break down in some remote areas where TEC infrastructures are unavailable (e.g., desert, isolated islands, and disaster-stricken areas). Driven by the ubiquitous connections and global-area coverage, space–air–ground-integrated networks (SAGINs) efficiently support seamless coverage and efficient resource management, and represent the next frontier for edge computing. In light of this, we first review the state-of-the-art edge computing research for SAGINs in this article. After discussing several existing orbital and aerial edge computing architectures, we propose a framework of edge computing-enabled SAGINs to support various Internet of Vehicles (EC-IoV) services for the vehicles in remote areas. The main objective of the framework is to minimize the task completion time and satellite resource usage. To this end, a preclassification scheme is presented to reduce the size of action space, and a deep imitation learning-driven offloading and caching algorithm is proposed to achieve real-time decision making. The simulation results show the effectiveness of our proposed scheme. Finally, we also discuss some technology challenges and future directions.

Directed Acyclic Graph (DAG) is a popular approach to achieve scalability of blockchain networks. Due to its high efficiency in data communication and great scalability, DAG has been widely adopted in many applications such as Internet of Things (IoT) and Decentralized Finance (DeFi). DAG-based blockchain, nevertheless, faces the key challenge of transaction inclusion collision due to the high concurrency and the network delay. Particularly, the transaction inclusion collision in DAG-based blockchain leads to the revenue and throughput dilemmas, which would greatly degrade the system performance. In this paper, we propose “TIPS”, the Transaction Inclusion Protocol with Signaling, which broadcasts a signal indicating the transactions in the block. We show that with the prompt broadcast of a signal, TIPS substantially reduces the transaction collision and thus resolves these dilemmas. Moreover, we show that TIPS can defend against both the denial-of-service and the delay-of-service attacks. We also conduct intensive experiments to demonstrate the superior performance of the proposed protocol.

To facilitate the emerging applications in 5G networks, mobile network operators will provide many network functions in terms of control and prediction. Recently, they have recognized the power of machine learning (ML) and started to explore its potential to facilitate those network functions. Nevertheless, the current ML models for network functions are often derived in an offline manner, which is inefficient due to the excessive overhead for transmitting a huge volume of dataset to remote ML training clouds and failing to provide the incremental learning capability for the continuous model updating. As an alternative solution, we proposeCocktail, an incremental learning framework within a reference 5G network architecture. To achieve cost efficiency while increasing trained model accuracy, an efficient online data scheduling policy is essential. To this end, we formulate an online data scheduling problem to optimize the framework cost while alleviating the data skew issue caused by the capacity heterogeneity of training workers from the long-term perspective. We exploit the stochastic gradient descent to devise an online asymptotically optimal algorithm, including two optimal policies based on novel graph constructions for skew-aware data collection and data training. Small-scale testbed and large-scale simulations validate the superior performance of our proposed framework.

Computation offloading is envisioned as a promising technique for prolonging the battery lives and enhancing the computation capability of mobile devices. In this paper, we study the task offloading and resource purchasing problems in an edge-cloud collaborative system. The purpose of this system is to minimize the cost of task offloading while ensuring that the tasks can be served before their maximum acceptable delays. Due to the uncertainty of both the task arrival rates and the prices of the computing resources, it is impossible to make an optimal decision online for a long-running time. Therefore, we propose a two-timescale Lyapunov optimization algorithm to overcome the uncertainty of the system’s future information and make the optimal decisions only based on the system’s current states. By purchasing computation resources in different timescales from the public cloud and making online decisions on where and how many requests should be offloaded, we can achieve an efficient outcome such that the system performance can approach the offline optimum without requiring a priori knowledge of system statistics. Rigorous theoretical analysis confirms the effectiveness of the proposed two-timescale Lyapunov optimization algorithm and extensive trace-driven experimental results show that the algorithm achieves outstanding performance gains over existing benchmarks.

Blockchain has attracted lots of attention in recent years. However, the performance of blockchain cannot meet the requirement of massive Internet of Things (IoT) devices. One of the important bottlenecks of blockchain is the limited computing resources on a single server while executing transactions. To address this issue, we propose Aeolus blockchain to achieve the distributed execution of blockchain transactions. There are two key challenges to achieving this for IoT blockchain: transaction structure and state consistency. Facing these challenges, we first propose a distributed blockchain transaction structure, which imports extra parameters to divide the transaction execution into different stages to enable distributed execution. Second, we propose distributed state update sharding, which equips each blockchain peer with its own master and shard servers. In this way, each blockchain peer can be considered as a cluster that distributes the transaction to shorten the processing time and reach the consensus finally. We implement Aeolus on Go-Ethereum to evaluate its feasibility, on a testbed including 132 cloud servers. Our system runs stably for more than 8 h under the workload of 190 000 000 real-world user transactions. Experimental results show the efficiency that Aeolus can achieve more than 100 000 transactions/s of blockchain transactions, which is 15.6 times the throughput of the original blockchain.

With the proliferation of online social networks, the identification or classification of real-life relationship between users has been very useful for many applications such as financial fraud detection. In real life, usually people with different relationships would present gifts with special meanings to each other on different dates. In many Asian cultures, especially in Chinese culture, the red packet is a traditional form of monetary gift. With the rapid development of the Internet, people gradually began to give electronic red packets instead of paper ones as the means of money gifting on social network platforms. As motivated, in this paper we advocate a novel approach that exploits users’ red packet interactions for users relationship identification on WeChat, one of the largest social platforms in China. Specifically, we analyze the WeChat red packets network, identify the real-life relationship types between users through mining the semantic information of the amount and sending time of each red packet. In order to better capture the red packet gifting behaviors between users for relationship identification, on one hand, we construct an Amount-Date Graph and apply the graph embedding method to learn embeddings of the amount and sending date of each red packet. On the other hand, we propose a novel sequential model, Cross & Attention Sequence Model (CASM), which explicitly learns the interactions between the latent semantic information of each red packet’s amount and sending date in the red packets sequence between two users. To validate our approach, we conduct comprehensive experiments on a real-world WeChat Users Red Packets dataset that involves 8 kinds of real-life relationships. The experiments show that our proposed approach performs significantly better than baselines and achieves 81.70 percent prediction accuracy.

Graph Neural Networks (GNNs) have led to state-of-the-art performance on a variety of machine learning tasks such as recommendation, node classification and link prediction. Graph neural network models generate node embeddings by merging nodes features with the aggregated neighboring nodes information. Most existing GNN models exploit a single type of aggregator (e.g., mean-pooling) to aggregate neighboring nodes information, and then add or concatenate the output of aggregator to the current representation vector of the center node. However, using only a single type of aggregator is difficult to capture the different aspects of neighboring information and the simple addition or concatenation update methods limit the expressive capability of GNNs. Not only that, existing supervised or semi-supervised GNN models are trained based on the loss function of the node label, which leads to the neglect of graph structure information. In this paper, we propose a novel graph neural network architecture, Graph Attention & Interaction Network (GAIN), for inductive learning on graphs. Unlike the previous GNN models that only utilize a single type of aggregation method, we use multiple types of aggregators to gather neighboring information in different aspects and integrate the outputs of these aggregators through the aggregator-level attention mechanism. Furthermore, we design a graph regularized loss to better capture the topological relationship of the nodes in the graph. Additionally, we first present the concept of graph feature interaction and propose a vector-wise explicit feature interaction mechanism to update the node embeddings. We conduct comprehensive experiments on two node-classification benchmarks and a real-world financial news dataset. The experiments demonstrate our GAIN model outperforms current state-of-the-art performances on all the tasks.

With the rapid development of mobile cellular technologies and the increasing popularity of mobile and Internet of Things (IoT) devices, timely mobile traffic forecasting with high accuracy becomes more and more critical for proactive network service provisioning and efficient network resource allocation in smart cities. Traditional traffic forecasting methods mostly rely on time series prediction techniques, which fail to capture the complicated dynamic nature and spatial relations of mobile traffic demand. In this paper, we propose a novel deep learning framework, graph attention spatial-temporal network (GASTN), for accurate citywide mobile traffic forecasting, which can capture not only local geographical dependency but also distant inter-region relationship when considering spatial factor. Specifically, GASTN considers spatial correlation through our constructed spatial relation graph and utilizes structural recurrent neural networks to model the global near-far spatial relationships as well as the temporal dependencies. In the framework of GASTN, two attention mechanisms are designed to integrate different effects in a holistic way. Besides, in order to further enhance the prediction performance, we propose a collaborative global-local learning strategy for the training of GASTN, which takes full advantage of the knowledge from both the global model and local models for individual regions and enhance the effectiveness of our model. Extensive experiments on a large-scale real-world mobile traffic dataset demonstrate that our GASTN model dramatically outperforms the state-of-the-art methods. And it reveals that a significant enhancement in the prediction performance of GASTN can be obtained by leveraging the collaborative global-local learning strategy.

In-home health monitoring has attracted great attention for the ageing population worldwide. With the abundant user health data accessed by Internet of Things (IoT) devices and recent development in machine learning, smart healthcare has seen many successful stories. However, existing approaches for in-home health monitoring do not pay sufficient attention to user data privacy and thus are far from being ready for large-scale practical deployment. In this paper, we propose FedHome, a novel cloud-edge based federated learning framework for in-home health monitoring, which learns a shared global model in the cloud from multiple homes at the network edges and achieves data privacy protection by keeping user data locally. To cope with the imbalanced and non-IID distribution inherent in user’s monitoring data, we design a generative convolutional autoencoder (GCAE), which aims to achieve accurate and personalized health monitoring by refining the model with a generated class-balanced dataset from user’s personal data. Besides, GCAE is lightweight to transfer between the cloud and edges, which is useful to reduce the communication cost of federated learning in FedHome. Extensive experiments based on realistic human activity recognition data traces corroborate that FedHome significantly outperforms existing widely-adopted methods.

Image-based three-dimensional (3D) reconstruction utilizes a set of photos to build 3D model and can be widely used in many emerging applications such as augmented reality (AR) and disaster recovery. Most of existing 3D reconstruction methods require a mobile user to walk around the target area and reconstruct objectives with a hand-held camera, which is inefficient and time-consuming. To meet the requirements of delay intensive and resource hungry applications in 5G, we propose an edge computing-based photo crowdsourcing (EC-PCS) framework in this paper. The main objective is to collect a set of representative photos from ubiquitous mobile and Internet of Things (IoT) devices at the network edge for real-time 3D model reconstruction, with network resource and monetary cost considerations. Specifically, we first propose a photo pricing mechanism by jointly considering their freshness, resolution and data size. Then, we design a novel photo selection scheme to dynamically select a set of photos with the required target coverage and the minimum monetary cost. We prove the NP-hardness of such problem, and develop an efficient greedy-based approximation algorithm to obtain a near-optimal solution. Moreover, an optimal network resource allocation scheme is presented, in order to minimize the maximum uploading delay of the selected photos to the edge server. Finally, a 3D reconstruction algorithm and a 3D model caching scheme are performed by the edge server in real time. Extensive experimental results based on real-world datasets demonstrate the superior performance of our EC-PCS system over the existing mechanisms.

Data crowdsourcing is a promising paradigm that leverages the “wisdom” of a potentially large crowd of “workers” in many application domains. Quality-aware crowdsourcing is beneficial as it makes use of workers’ data quality to perform task allocation and data aggregation. However, a worker’s quality and data can be her private information that she may have incentive to misreport to the crowdsourcing requester. Moreover, a worker’s quality and data can depend on her sensitive information (e.g., location), which can be inferred from the outcomes of task allocation and data aggregation by an adversary. In this paper, we devise Privacy-preserving crowdsourcing mechanisms for truthful Data Quality Elicitation (PDQE). In these mechanisms, we design differentially private task allocation and data aggregation algorithms to prevent the inference of a worker’s quality and data from the outcomes of these algorithms. In the meantime, the mechanisms also incentivize workers to truthfully report their quality and data and make desired efforts. We first focus on the mechanisms for a single task (S-PDQE) and then extend it to the case of multiple tasks (M-PDQE). We further show that both the mechanisms achieve a bounded performance gap compared to the optimal strategy. We evaluate the proposed mechanisms using simulations based on real-world data, which corroborate their highly-desired properties on truthful data quality elicitation, data accuracy and privacy protection.

Numerous applications in 3D underwater sensor networks (UWSNs), such as pollution detection, disaster prevention, animal monitoring, navigation assistance, and submarines tracking, heavily rely on accurate localization techniques. However, due to the limited batteries of sensor nodes and the difficulty for energy harvesting in UWSNs, it is challenging to localize sensor nodes successfully within a short sensor node lifetime in an unspecified underwater environment. Therefore, we propose the Adaptive Energy-Efficient Localization Algorithm (Adaptive EELA) to enable energy-efficient node localization while adapting to the dynamic environment changes. Adaptive EELA takes a fuzzy game-theoretic approach, whereby the Stackelberg game is used to model the interactions among sensor and anchor nodes in UWSNs and employs the adaptive neuro-fuzzy method to set the appropriate utility functions. We prove that a socially optimal Stackelberg–Nash equilibrium is achieved in Adaptive EELA. Through extensive numerical simulations under various environmental scenarios, the evaluation results show that our proposed algorithm accomplishes a significant energy reduction, e.g., 66% lower compared to baselines, while achieving a desired performance level in terms of localization coverage, error, and delay.

In an online social network, users exhibit personal information to enjoy social interaction. The social network provider (SNP) exploits users’ information for revenue generation through targeted advertisement, in which the SNP presents advertisements to proper users effectively. Therefore, an advertiser is more willing to pay for targeted advertisement to promote his product. However, the over-exploitation of users’ information would invade users’ privacy, which would negatively impact users’ social activeness. Motivated by this, we study the privacy policy (policies) of the SNP(s) with targeted advertisement, in both monopoly and duopoly markets. We characterize the privacy policy in terms of the fraction of users’ information that the provider should exploit, and formulate the interactions among users, advertiser, and SNP(s) as a three-stage Stackelberg game. By leveraging the model’s supermodularity property, we prove the threshold structure of users’ equilibrium information levels. We discover the overall information that can be exploited by an SNP is non-monotonic in the exploitation fraction. Monopoly (one SNP) study shows our proposed optimal privacy policy helps the SNP earn even more advertisement revenue than full exploitation policy does. The situation of the duopoly market is much more complicated. In that case, if the service quality gap between the two SNPs is large, the stronger SNP will choose a conservative privacy protection policy that drives the other SNP out of the market. However, if the service quality gap is small and the advertisement revenue is promising, the stronger SNP would choose an aggressive policy to exploit the advertisement revenue and both SNPs will have positive market shares.

With the growth of participating clients, the centralized parameter server (PS) will seriously limit the scale and efficiency of Federated Learning (FL). A straightforward approach to scale up the FL system is to construct a Parallel FL (PFL) system with multiple parallel PSes. However, it is unclear whether PFL can really accelerate FL or reduce the training time of FL. Even if the answer is yes, it is non-trivial to design a highly efficient parameter average algorithm for a PFL system. In this paper, we propose a completely parallelizable FL algorithm called P-FedAvg under the PFL architecture. P-FedAvg extends the well-known FedAvg algorithm by allowing multiple PSes to cooperate and train a learning model together. In P-FedAvg, each PS is only responsible for a fraction of total clients, but PSes can mix model parameters in a dedicatedly designed way so that the FL model can well converge. Different from heuristic-based algorithms, P-FedAvg is with theoretical guarantees. To be rigorous, we theoretically analyze the convergence rate of P-FedAvg in terms of the number of conducted iterations, the communication cost of each global iteration and the optimal weights for each PS to mix parameters with its neighbors. Based on theoretical analysis, we conduct a case study on five typical overlay topolgoies formed by PSes to further examine the communication efficiency under different topologies, and investigate how the overlay topology affects the convergence rate, communication cost and robustness of a PFL system. Lastly, we perform extensive experiments with real datasets to verify our analysis and demonstrate that P-FedAvg can significantly speed up FL than traditional FedAvg and other competitive baselines. We believe that our work can help to lay a theoretical foundation for building more efficient PFL systems.

Precise citywide mobile traffic prediction is of great significance for intelligent network planning and proactive service provisioning. Current traffic prediction approaches mainly focus on training a well-performed model for the cities with a large amount of mobile traffic data. However, for the cities with scarce data, the prediction performance will be greatly limited. To tackle this problem, in this paper we propose a novel cross-city deep transfer learning framework named CCTP for citywide mobile traffic prediction in cities with data scarcity. Specifically, we first present a novel spatial-temporal learning model and pre-train the model by abundant data of a source city to obtain prior knowledge of mobile traffic dynamics. We then devise an efficient generative adversarial network (GAN) based cross-domain adapter for distribution alignment between target data and source data. To deal with data scarcity issue in some clusters of target city, we further design an inter-cluster transfer learning strategy for performance enhancement. Extensive experiments conducted on real-world mobile traffic datasets demonstrate that our proposed CCTP framework can achieve superior performance in citywide mobile traffic prediction with data scarcity.

Federated learning (FL) emerges as a promising paradigm to enable a federation of clients to train a machine learning model in a privacy-preserving manner. Most existing works assumed that the central parameter server (PS) determines the participation of clients implying that clients cannot make autonomous participation decisions. The above assumption is unrealistic because the participation in FL training may incur various cost and clients also have strong desire to be rewarded for participation. To address this problem, we design a novel autonomous client participation scheme to incentivize clients. Specifically, the PS provides a certain reward shared among participating clients for each training round. Clients decide whether to participate each FL training round or not based on their own utilities (i.e., reward minus cost). The process can be modeled as a minority game (MG) with incomplete information and clients end up in the minority side win after each training round because the reward of each participating client may not cover its cost if too many clients participate and vice verse. The challenge of autonomous participation schemes lies in lowering thevolatilityof participating clients in each round due to the lack of coordination among clients. Through solid analysis, we prove that: 1) The volatility of participating clients in each round is very high under the standard MG scheme. 2) The volatility of participating clients can be reduced significantly under the stochastic MG scheme. 3) A coalition based MG is proposed, which can further reduce the volatility in each round. By conducting extensive experiments in real settings, we demonstrate that the stochastic MG-based scheme outperforms other state-of-the-art algorithms in terms of utility and volatility, and the coalition MG-based client participation scheme can further boost the utility by 39%-48% and reduce the volatility by 51%–100%. Moreover, our algorithms can achieve almost the same model accuracy as that obtained by centralized client participation algorithms.

As mobile edge computing (MEC) hosting applications at the network edge with limited capacities, service providers are facing the new challenge of how to make full use of the scarce edge resources to maximize the system performance. Accommodating this challenge requires careful application placement and request routing to coordinate diverse MEC nodes. However, frequent application re-placement would greatly increase the system reconfiguration cost, indicating a performance-cost trade-off. In response, in this paper, we study the problem of joint optimization on application placement and request routing to maximize the system performance, under a long-term budget of the application reconfiguration cost. Solving this problem is non-trivial since the long-term budget is coupled with the future system states (e.g., user request arrivals) that are typically unpredictable. To address this challenge, we first advocate an approximated dynamic optimization framework to decompose the long-term optimization problem into a series of one-shot problems which do not require the future system states. Moreover, since the decomposed problem is a mixed integer linear program (MILP) which is proven to be NP-hard, we then devise an efficient dependent rounding based approximation algorithm, which can achieve the near-optimal performance in a fast manner. Both rigorous theoretical analysis and extensive trace-driven evaluations demonstrate the proposed framework can achieve superior performance gain over existing schemes.

With the rapid development of 5G networks, a great amount of Internet of Things (IoT) devices are connected to the Internet. Most of these devices are cost limited and thus are easily compromised by attackers to launch distributed denial of service (DDoS) attacks. The traditional DDoS defense methods at server side can not adapt to this new challenge, thus access-side DDoS detection architecture is urgently needed. In this paper, we propose a deep learning (DL) based IoT device classification method to support fine-grained behavior modeling of malicious traffic and thus enable access-side DDoS detection. Different from traditional studies based on machine learning (ML) which need expertise feature engineering, we propose a time characteristics extraction method based on 1-D convolutional neural network to capture high level time series features automatically for better classification performance. To avoid the feature loss problem, we propose a feature enhancement method based on residual connection module. Experimental results verify the effectiveness of our method, which offers a meaningful gain in terms of both accuracy and macro F1 score over existing approaches.

Grant-free random access has the potential to sup-port massive connectivity in Internet of Things (IoT) networks, where joint activity detection and channel estimation (JADCE) is a key issue that needs to be tackled. The existing methods for JADCE usually suffer from one of the following limitations: high computational complexity, ineffective in inducing sparsity, and incapable of handling complex matrix estimation. To mitigate all the aforementioned limitations, we in this paper develop an effective unfolding neural network framework built upon the proximal operator method to tackle the JADCE problem in IoT networks, where the base station is equipped with multiple antennas. Specifically, the JADCE problem is formulated as a group-sparse-matrix estimation problem, which is regularized by non-convex minimax concave penalty (MCP). This problem can be iteratively solved by using the proximal operator method, based on which we develop a unfolding neural network structure by parameterizing the algorithmic iterations. By further exploiting the coupling structure among the training parameters as well as the analytical computation, we develop two additional unfolding structures to reduce the training complexity. We prove that the proposed algorithm achieves a linear convergence rate. Results show that our proposed three unfolding structures not only achieve a faster convergence rate but also obtain a higher estimation accuracy than the baseline methods.

With the rapid development of the Internet of Things (IoT), more and more smart devices are connected to the Internet, implementing Deep Neural Network (DNN) models on edges for collaborative inference via device-edge synergy has become a feasible method for improving application performance in many scenarios. However, when faced with the task stream scenario with latency guarantee such as video surveillance and industrial production line, we need adaptive edge intelligence to make adjustments in real-time according to the changes of the task stream. There are many adaptive edge intelligence technologies in the existing works, such as early-exit mechanism and model selection, but they don’t take the requirements of the task stream scenario into consideration. In this paper, we propose a device-edge collaborative inference system based on the early-exit mechanism to solve the problem of adaptive edge intelligence in the task stream scenario. Then, we design an offline dynamic programming (DP) algorithm and an online deep reinforcement learning (DRL) algorithm to dynamically select the exit point and partition point of the branchy model in the task stream, which aims to balance the number of tasks accomplished and task inference accuracy in the system. Experimental results show that the DRL algorithm can achieve performance close to that of the DP algorithm in the task stream scenario.

With the growth of participating clients, the centralized parameter server (PS) will seriously limit the scale and efficiency of Federated Learning (FL). A straightforward approach to scale up the FL system is to construct a Parallel FL (PFL) system with multiple PSes. However, it is unclear whether PFL can really achieve a faster convergence rate or not. Even if the answer is yes, it is non-trivial to design a highly efficient parameter average algorithm for a PFL system. In this paper, we propose a completely parallelizable FL algorithm called P-FedAvg under the PFL architecture. P-FedAvg extends the well-known FedAvg algorithm by allowing multiple PSes to cooperate and train a learning model together. In P-FedAvg, each PS is only responsible for a fraction of total clients, but PSes can mix model parameters in a dedicatedly designed way so that the FL model can well converge. Different from heuristic-based algorithms, P-FedAvg is with theoretical guarantees. To be rigorous, we conduct theoretical analysis on the convergence rate of P-FedAvg, and derive the optimal weights for each PS to mix parameters with its neighbors. We also examine how the overlay topology formed by PSes affects the convergence rate and robustness of a PFL system. Lastly, we perform extensive experiments with real datasets to verify our analysis and demonstrate that P-FedAvg can significantly improve convergence rates than traditional FedAvg and other competitive baselines. We believe that our work can help to lay a theoretical foundation for building more efficient PFL systems.

Sharding is one of the most promising and practical methods to achieve horizontal scalability of blockchain networks. However, the increasing number of cross-shard transactions in blockchain sharding protocols may degrade the system throughput. In this paper, we investigate how to distribute users properly in the shard-based blockchains to boost the system transaction performance. We first build an open Jackson queueing network model to capture users' transaction dynamics on shards. Then we cast users' interactions as a shard-based blockchain game, wherein each user aims to minimize its transaction confirmation time and transaction fee. We investigate the equilibrium of the game, and design a polynomial-time algorithm to find efficient equilibria with good system performance. We further design a novel sharding protocol with dynamic user distribution for the permissionless blockchain, and the protocol can maintain good performance in long-term dynamic environment. Extensive numerical results using realistic blockchain transaction data demonstrate that the proposed algorithm and the designed protocol can achieve superior performance for shard-based blockchains.

As edge computing provides crucial support for delay-sensitive and computation-intensive applications, many business entities deploy their own edge servers to compete for users, which forms multi-server edge computing networks. However, no prior work studies the competition among heterogeneous edge servers and how the competition affects users’ selfish computation offloading behaviors in such a network from an economic perspective. In this paper, we model the interactions between edge servers and users as a two-stage game. In Stage I, edge servers with heterogeneous marginal costs set their service prices to compete for users, and in Stage II, each user selfishly offloads its task to one of the edge servers or the remote cloud. Analyzing the equilibrium of the two-stage game is challenging due to edge servers’ heterogeneity and the congestion effect caused by resource sharing among users. We first prove that in Stage II, users’ selfish computation offloading game is a potential game and admits a unique Nash equilibrium (NE), for which we derive the explicit expression. We then analyze edge servers’ price competition game in Stage I and characterize the conditions for the uniqueness of the NE. We show that at equilibrium, users only choose low-priced edge servers, and hence edge servers with low marginal costs can win the price competition, which reflects the improvement of economic efficiency in competitive markets. Moreover, it is surprising that the equilibrium prices do not monotonically increase with the task execution delay. This is because a long execution delay gives a chance to edge servers with high marginal costs to win the competition, which results in more fierce competition among edge servers.

The freshness of status information is of great importance for time-critical Internet-of-Things (IoT) applications. A metric measuring status freshness is the Age of Information (AoI), which captures the time elapsed from the status being generated at the source node (e.g., a sensor) to the latest status update. However, in intelligent IoT applications such as video surveillance, the status information is revealed after some computation-intensive and time-consuming data processing operations, which would affect the status freshness. In this article, we propose a novel metric, Age of Processing (AoP), to quantify such status freshness, which captures the time elapsed of the newest received processed status data since it is generated. Compared with AoI, AoP further takes the data processing time into account. Since an IoT device has limited computation and energy resources, the IoT device can choose to offload the data processing to the nearby edge server under constrained status sampling frequency. We aim to minimize theaverageAoP in a long-term process by jointly optimizing the status sampling frequency and processing offloading policy. We first formulate this online problem as an infinite-horizon constrained Markov decision process (CMDP) with an average reward criterion. We then transform the CMDP problem into an unconstrained Markov decision process (MDP) by leveraging a Lagrangian method, and accordingly propose a Lagrangian transformation framework for the original CMDP problem. Furthermore, we integrate the framework with a perturbation-based refinement mechanism for achieving the optimal policy of the CMDP problem. Our investigation shows that to minimize the average AoP: 1) for processing offloading: the policy exploits good channel state to offload processing to the edge server and 2) for status sampling: the waiting time presents a threshold structure. Extensive numerical evaluations show that the proposed algorithm outperforms the benchmarks, with an average AoP reduction up to 30%.

Mobile-edge computing (MEC) has emerged as a promising supporting architecture providing a variety of resources to the network edge, thus acting as an enabler for edge intelligence services empowering massive mobile and Internet-of-Things (IoT) devices with artificial intelligence (AI) capability. With the assistance of edge servers, user equipments (UEs) are able to run deep neural network (DNN)-based AI applications, which are generally resource hungry and computation intensive such that an individual UE can hardly afford by itself in real time. However, the resources in each individual edge server are typically limited. Therefore, any resource optimization involving edge servers is by nature a resource-constrained optimization problem and needs to be tackled in such a realistic context. Motivated by this observation, we investigate the optimization problem of DNN partitioning (an emerging DNN offloading scheme) in a realistic multiuser resource-constrained condition that rarely considered in previous works. Despite the extremely large solution space, we reveal several properties of this specific optimization problem of joint multi-UE DNN partitioning and computational resource allocation. We propose an algorithm called iterative alternating optimization (IAO) that can achieve the optimal solution in polynomial time. In addition, we present a rigorous theoretic analysis of our algorithm in terms of time complexity and performance under realistic estimation error. Moreover, we build a prototype that implements our framework and conducts extensive experiments using realistic DNN models, whose results demonstrate its effectiveness and efficiency.

Cloud radio access network (C-RAN) is a promising 5G network architecture by establishing baseband units (BBU) pools to perform baseband processing functionalities and deploying remote radio heads (RRHs) for wireless signal transmission and reception. Mobile-edge computing (MEC) offers a way to shorten the service delay by building small-scale cloud infrastructures at the network edge. By co-locating the BBU pool with edge cloud at the so-called BBU node, we can take full advantages of C-RAN and MEC for better spectrum utilization and delay-guaranteed services. In this article, we first study how to allocate each user's task to the BBU node and find the path from his/her accessing RRH node to the BBU node such that the maximum service delay among all the requests is minimized. We then consider this problem with survivability concerns, which is to use both primary and backup BBU nodes to issue the request such that the primary path and backup path are link disjoint. We analyze the complexities of these two problems and prove they are NP-hard in general. Subsequently, we devise a randomized approximation algorithm and an efficient heuristic to solve the considered problems, respectively. The simulation results show that the proposed algorithms outperform two benchmark heuristics in terms of acceptance ratio and maximum service delay.

Recently, smart cities, healthcare system, and smart vehicles have raised challenges on the capability and connectivity of state-of-the-art Internet-of-Things (IoT) devices, especially for the devices in hotspots area. Multiaccess edge computing (MEC) can enhance the ability of emerging resource-intensive IoT applications and has attracted much attention. However, due to the time-varying network environments, as well as the heterogeneous resources of network devices, it is hard to achieve stable, reliable, and real-time interactions between edge devices and their serving edge servers, especially in the 5G ultradense network (UDN) scenarios. Ultradense edge computing (UDEC) has the potential to fill this gap, especially in the 5G era, but it still faces challenges in its current solutions, such as the lack of: 1) efficient utilization of multiple 5G resources (e.g., computation, communication, storage, and service resources); 2) low overhead offloading decision making and resource allocation strategies; and 3) privacy and security protection schemes. Thus, we first propose an intelligent UDEC (I-UDEC) framework, which integrates blockchain and artificial intelligence (AI) into 5G UDEC networks. Then, in order to achieve real-time and low overhead computation offloading decisions and resource allocation strategies, we design a novel two-timescale deep reinforcement learning (2Ts-DRL) approach, consisting of a fast-timescale and a slow-timescale learning process, respectively. The primary objective is to minimize the total offloading delay and network resource usage by jointly optimizing computation offloading, resource allocation, and service caching placement. We also leverage federated learning (FL) to train the 2Ts-DRL model in a distributed manner, aiming to protect the edge devices' data privacy. Simulation results corroborate the effectiveness of both the 2Ts-DRL and FL in the I-UDEC framework and prove that our proposed algorithm can reduce task execution time up to 31.87%.

Mobile Edge Computing (MEC) offers a way to shorten the cloud servicing delay by building the small-scale cloud infrastructures at the network edge, which are in close proximity to the end users. Moreover, Network Function Virtualization (NFV) has been an emerging technology that transforms from traditional dedicated hardware implementations to software instances running in a virtualized environment. In NFV, the requested service is implemented by a sequence of Virtual Network Functions (VNF) that can run on generic servers by leveraging the virtualization technology. Service Function Chaining (SFC) is defined as a chain-ordered set of placed VNFs that handles the traffic of the delivery and control of a specific application. NFV therefore allows to allocate network resources in a more scalable and elastic manner, offer a more efficient and agile management and operation mechanism for network functions and hence can largely reduce the overall costs in MEC. In this paper, we study the problem of how to place VNFs on edge and public clouds and route the traffic among adjacent VNF pairs, such that the maximum link load ratio is minimized and each user's requested delay is satisfied. We consider this problem for both totally ordered SFCs and partially ordered SFCs. We prove that this problem is NP-hard, even for the special case when only one VNF is requested. We subsequently propose an efficient randomized rounding approximation algorithm to solve this problem. Extensive simulation results show that the proposed approximation algorithm can achieve close-to-optimal performance in terms of acceptance ratio and maximum link load ratio.

To meet the ever increasing mobile traffic demand in 5G era, base stations (BSs) have been densely deployed in radio access networks (RANs) to increase the network coverage and capacity. However, as the high density of BSs is designed to accommodate peak traffic, it would consume an unnecessarily large amount of energy if BSs are on during off-peak time. To save the energy consumption of cellular networks, an effective way is to deactivate some idle base stations that do not serve any traffic demand. In this paper, we develop a traffic-aware dynamic BS sleep control framework, named DeepBSC, which presents a novel data-driven learning approach to determine the BS active/sleep modes while meeting lower energy consumption and satisfactory Quality of Service (QoS) requirements. Specifically, the traffic demands are predicted by the proposed GS-STN model, which leverages the geographical and semantic spatial-temporal correlations of mobile traffic. With accurate mobile traffic forecasting, the BS sleep control problem is cast as a Markov Decision Process that is solved by Actor-Critic reinforcement learning methods. To reduce the variance of cost estimation in the dynamic environment, we propose a benchmark transformation method that provides robust performance indicator for policy update. To expedite the training process, we adopt a Deep Deterministic Policy Gradient (DDPG) approach, together with an explorer network, which can strengthen the exploration further. Extensive experiments with a real-world dataset corroborate that our proposed framework significantly outperforms the existing methods.

Recent advances in artificial intelligence have driven increasing intelligent applications at the network edge, such as smart home, smart factory, and smart city. To deploy computationally intensive Deep Neural Networks (DNNs) on resource-constrained edge devices, traditional approaches have relied on either offloading workload to the remote cloud or optimizing computation at the end device locally. However, the cloud-assisted approaches suffer from the unreliable and delay-significant wide-area network, and the local computing approaches are limited by the constrained computing capability. Towards high-performance edge intelligence, the cooperative execution mechanism offers a new paradigm, which has attracted growing research interest recently. In this paper, we propose CoEdge, a distributed DNN computing system that orchestrates cooperative DNN inference over heterogeneous edge devices. CoEdge utilizes available computation and communication resources at the edge and dynamically partitions the DNN inference workload adaptive to devices' computing capabilities and network conditions. Experimental evaluations based on a realistic prototype show that CoEdge outperforms status-quo approaches in saving energy with close inference latency, achieving up to 25.5% ~ 66.9% energy reduction for four widely-adopted CNN models.

Distributed Reflection Denial of Service (DRDoS) attack has become one of the most serious threats to Internet security. With the ongoing development of 5G, a massive number of insecure Internet of Things (IoT) devices are connected to the Internet, which brings great challenges to defend against DRDoS attacks. To overcome these challenges, we extend the User Plane Function (UPF) of 5G core network, and propose a new framework accordingly for source IP address validation, so as to suppress the source IP address spoofing behaviors of DRDoS attackers. Under this framework, the packet inspection rate (PIR), i.e., the inspection probability of each packet, is crucial to simplify the validation complexity. To unveil the optimal PIR, we establish a two-player game which models the IP address spoofing and detection behaviors. Analysis on the formulated game implies a lower bound of sufficient PIR, which may be used to set PIR in practice. Simulation results show that the proposed method can efficiently deter IP spoofing behaviors. Thereby the derived PIR could achieve low-cost and effective defense of DRDoS.

The link flooding attack (LFA) arises as a new class of Distributed Denial of Service (DDoS) attacks in recent years. By aggregating low-rate protocol-conforming traffic to congest selected links, LFAs can degrade the connectivity of target servers indirectly. Due to the fast proliferation of insecure Internet of Things (IoT) devices, the deployment of botnets is getting easier, which dramatically increases the risk of LFAs. Since the attacking traffic may not reach the victims directly and seems to be legitimate, LFAs are extremely difficult to detect and defend using traditional methods. In this work, we model the interaction between the LFA attacker and the defender as an extensive form game with incomplete information. By using action space compression and the divide and conquer method, we analyze the Nash equilibrium of the subgame on each link, which reveals the rational behaviors of attackers and the optimal strategies of defenders. Furthermore, we concretely expound how to adopt local optimal strategies in the Internet-wide scenario. Experimental results show the effectiveness and robustness of our proposed decision-making method in explicit LFA defending scenarios.

With the growing size of the cloud data center, the high scheduling efficiency over massive-scale cloud servers is hard to achieve, particularly when the scheduler requires the full real-time cloud resource information for decision making. Moreover, most data centers only run latency-critical online services, resulting in low resource utilization. To solve these problems, we propose a Compressive Sensing based Predictive Online Scheduling (CSPOS) algorithm. To mitigate the bottleneck of transferring massive resource information of all cloud servers to the scheduler, we propose to transfer sampled data from a small subset of servers to the scheduler and recover the full cloud resource information by compressive sensing. We then propose a predictive online learning algorithm that efficiently colocates the online services and batch jobs, in order to boost the resource utilization of the data center. Our experiments show that the CSPOS model achieves outstanding scheduling efficiency under various settings and is able to greatly increase the resource usage of a data center. We also illustrate that the running time of the CSPOS model is very small and has negligible effects on the scheduling system.

In an online social network, users exhibit personal information to enjoy social interaction. The social network provider (SNP) exploits users' information for revenue generation through targeted advertising. The SNP can present ads to proper users efficiently. Therefore, an advertiser is more willing to pay for targeted advertising. However, the over-exploitation of users' information would invade users' privacy, which would negatively impact users' social activeness. Motivated by this, we study the optimal privacy policy of the SNP with targeted advertising business. We characterize the privacy policy in terms of the fraction of users' information that the provider should exploit, and formulate the interactions among users, advertiser, and SNP as a three-stage Stackelberg game. By carefully leveraging supermodularity property, we reveal from the equilibrium analysis that higher information exploitation will discourage users from exhibiting information, lowering the overall amount of exploited information and harming advertising revenue. We further characterize the optimal privacy policy based on the connection between users' information levels and privacy policy. Numerical results reveal some useful insights that the optimal policy can well balance the users' trade-off between social benefit and privacy loss.

Recent years have witnessed the proliferation of mobile computing and the Internet of Things (IoT), in which billions of mobile and IoT devices are connected to the Internet, generating zillions bytes of data at the network edge. However, it is challenging and infeasible to transfer and process zillions bytes of data using the current cloud-device architecture, due to bandwidth constraints of networks, potentially uncontrollable latency of cloud services, and privacy concerns while collecting data from IoT devices. To tackle these challenges, edge computing, an emerging computing paradigm, has received a tremendous amount of interest. By pushing data storage, computing, and controls closer to the network edge, edge computing has been widely recognized as a promising solution to meet the requirements of low latency, high scalability, and energy efficiency, as well as to mitigate the network traffic burdens. However, with the emergence of diverse IoT applications (e.g., smart city, industrial automation, and connected car), it becomes challenging for edge computing to deal with these heterogeneous IoT environments.

Currently, data sharing for the Internet of Vehicles (IoV) applications has drawn much attention in the framework of developing smart cities and smart transportation. A critical challenge for data sharing is to incentivize users to participate in collecting and sharing data. The traditional incentive mechanism of crowdsourcing is not practical for IoV because of its trust issues. Although blockchain technology has been introduced to address trust issues and security challenges, ensuring trust in off-chain data for the blockchain-based approaches is still an open issue. In this article, we propose a quality-driven auction-based incentive mechanism based on a consortium blockchain that guarantees trust in both on-chain data and off-chain data. We first introduce a consortium blockchain that is used as an open and distributed hyperledger to address the security issue of on-chain data. Then, we formulate the problem as a reverse auction in which the platform acts as an auctioneer that purchases data from users. By utilizing a data quality-driven auction model, the evaluated data quality via expectation maximization is used to ensure the trust in off-chain data. The quality-driven, auction-based incentive mechanism can obtain the high-quality data and optimal social welfare with low social cost. Otherwise, we design a smart contract to perform the data sharing automatically. Finally, the extensive simulations show that our proposed algorithm achieves maximum social welfare, outperforms other solutions, and scales well when the number of users or tasks increase. Moreover, the performance of the smart contract shows its low computing cost.

A key challenge for autonomous driving is to process a massive amount of sensor data and make safe and reliable decisions in real time. However, autonomous vehicles often have insufficient onboard resources to provide the required computation capacity. To address this problem, this article advocates a novel approach to offload computation-intensive autonomous driving services to roadside units and cloud for swift executions. Our approach combines an integer linear programming (ILP) formulation for offline optimization of the scheduling strategy and a fast heuristics algorithm for online adaptation. We verify our technique with both synthetic task graphs and real-world deployment. The experimental results show that our approach can improve system performance effectively.

With the great advance in the Internet-of-Things (IoT) sector, the recent years have witnessed an unprecedented wave of the proliferation of heterogeneous IoT devices and applications. Among them, some have stringent hard deadlines which can only be satisfied by the emerging paradigm of mobile-edge computing (MEC), while the others may pose elastic soft deadlines which can be flexibly fulfilled by cloud computing. However, with the presence of both temporal and spatial diversities of the resource cost of MEC and cloud, it remains a practical challenge how to efficiently provision the MEC and cloud resource to minimize the long-term operational cost, while still guaranteeing both hard and soft deadlines for heterogeneous IoT applications. To navigate such an inherent performance-cost tradeoff, an efficient online cloud-edge resource provisioning framework is proposed, based on the delay-aware Lyapunov optimization technique. Without requiring a priori knowledge of the statistics of the cloud-edge system, the proposed framework allows to make online greedy decisions on how much MEC and cloud resources to be provisioned to heterogeneous IoT applications. Through rigorous theoretical analysis, we prove that without violating both the hard and soft deadlines of heterogeneous IoT applications, the long-term operational cost can be pushed arbitrarily close to the offline optimum. With extensive evaluations driven by realistic traffic and cost traces, we empirically demonstrate the cost efficiency of the proposed cloud-edge resource provisioning framework.

Online social networks (OSNs) are emerging as the most popular mainstream platform for content cascade diffusion. In order to provide satisfactory quality of experience (QoE) for users in OSNs, much research dedicates to proactive content placement by using the propagation pattern, user's personal profiles and social relationships in open social network scenarios (e.g., Twitter and Weibo). In this paper, we take a new direction of popularity-aware content placement in a closed social network (e.g., WeChat Moment) where user's privacy is highly enhanced. We propose a novel data-driven holistic deep learning framework, namely DeepCP, for joint diffusion-aware cascade prediction and autonomous content placement without utilizing users' personal and social information. We first devise a time-window LSTM model for content popularity prediction and cascade geo-distribution estimation. Accordingly, we further propose a novel autonomous content placement mechanism CP-GAN which adopts the generative adversarial network (GAN) for agile placement decision making to reduce the content access latency and enhance users' QoE. We conduct extensive experiments using cascade diffusion traces in WeChat Moment (WM). Evaluation results corroborate that the proposed DeepCP framework can predict the content popularity with a high accuracy, generate efficient placement decision in a real-time manner, and achieve significant content access latency reduction over existing schemes.

The explosive growth of traffic-consumption by mobile devices is leading to severe cellular network congestion, which is posing challenges for Internet Service Providers (ISPs) to provide good quality services with limited cellular capacity and impacting the user's experience. Data pricing has been proven to be an effective way to enhance both the service quality and ISP's profit. However, traditional data pricing schemes do not consider the real Mobile Application (App) Usage Patterns (MAUPs) among large scale cellular networks. In this paper, MAUPs aware smart data pricing scheme is proposed. In our work, we firstly extract and model the users' app usage behaviors of approximately 9,600 cellular towers as two-dimensional MAUPs (time, app category). Then 7 distinct derived MAUPs are considered to be incorporated into the user satisfaction model and ISP's profit model. The performance of our proposal is evaluated and verified by numerical experiments from the aspects of ISP's profit, consumption surplus, capacity utilization and traffic efficiency. The MAUPs based pricing scheme can be periodically updated according to the operational conditions and therefore significantly instructive for ISPs.

We study a problem of privacy-preserving mechanism design. A data collector wants to obtain data from individuals to perform some computations. To relieve the privacy threat to the contributors, the data collector adopts a privacy-preserving mechanism by adding random noise to the computation result, at the cost of reduced accuracy. Individuals decide whether to contribute data when faced with the privacy issue. Due to the intrinsic uncertainty in privacy protection, we model individuals' privacy-related decision using Prospect Theory. Such a theory more accurately models individuals' behavior under uncertainty than the traditional expected utility theory, whose prediction always deviates from practical human behavior. We show that the data collector's utility maximization problem involves a polynomial of high and fractional order, the root of which is difficult to compute analytically. We get around this issue by considering a large population approximation, and obtain a closed-form solution that well approximates the precise solution. We discover that the data collector who considers the more realistic Prospect Theory based individual decision modeling would adopt a more conservative privacy-preserving mechanism, compared with the case based on the expected utility theory modeling. We also study the impact of Prospect Theory parameters, and concludes that more loss-averse or risk-seeking individuals will trigger a more conservative mechanism. When individuals have different Prospect Theory parameters, simulations demonstrate that the privacy protection first becomes stronger and then becomes weaker as the heterogeneity increases from a low value to a high one.

We study a privacy-preserving data collection problem by considering individuals' data correlation and social relationship. A data collector gathers data from some data reporters to perform certain analysis with a privacy-preserving mechanism. Due to the data correlation, the analysis will cause privacy leakage not only to the data reporters but also to those individuals who do not report data. Owing to the social relationship among them, the data reporters would consider the possibility of adding some random noise to the reported data to reduce the privacy leakage. The privacy loss of the individuals (both data reporters and non-reporters) depend on all the data reporters' strategies, which naturally leads to a game theoretical analysis. A key result shows that the data reporters can be ordered based on their levels of joint considerations of social relationship and data correlation, and at the Nash Equilibrium of the game at most one data reporter with the most significant consideration may add noise to the reported data. We design an efficient algorithm for the data collector to construct the data reporter set, and derive the optimal privacy-preserving mechanism to ensure all the data reporters' truthful reporting. We conduct extensive simulations with the Facebook social data to demonstrate some insights: It is optimal for the data collector to adopt a more conservative mechanism when the data correlation or the social relationship is stronger. Compared with the data correlation information, the social network information plays a more critical role in the data collector's utility maximization problem.

Task offloading is one of key operations in edge computing, which is essential for reducing the latency of task processing and boosting the capacity of end devices. However, the heterogeneity among tasks generated by various users makes it challenging to design efficient task offloading algorithms. In addition, the assumption of complete information for offloading decision-making does not always hold in a distributed edge computing environment. In this article, we formulate the problem of heterogeneous task offloading in a distributed environment as a minority game (MG), in which each player must make decisions independently in each turn and the players who end up on the minority side win. The multi-player MG incentivizes players to cooperate with each other in the scenarios with incomplete information, where players don't have full information about other players (e.g., the number of tasks, the required resources). To address the challenges incurred by task heterogeneity and the divergence of naive MG approaches, we propose an MG based scheme, in which tasks are divided into subtasks and instructed to form into a set of groups as possible, and the left ones are scheduled to perform decision adjustment in a probabilistic manner. We prove that our proposed algorithm can converge to a near-optimal point, and also investigate its stability and price of anarchy in terms of task processing time. Finally, we conduct a series of simulations to evaluate the effectiveness of our proposed scheme and the results indicate that our scheme can achieve around 30% reduction of task processing time compared with other approaches. Moreover, our proposed scheme can converge to a near-optimal point, which cannot be guaranteed by naive MG approaches.

As a key technology of enabling Artificial Intelligence (AI) applications in 5G era, Deep Neural Networks (DNNs) have quickly attracted widespread attention. However, it is challenging to run computation-intensive DNN-based tasks on mobile devices due to the limited computation resources. What’s worse, traditional cloud-assisted DNN inference is heavily hindered by the significant wide-area network latency, leading to poor real-time performance as well as low quality of user experience. To address these challenges, in this paper, we proposeEdgent, a framework that leverages edge computing for DNN collaborative inference through device-edge synergy.Edgentexploits two design knobs: (1) DNN partitioning that adaptively partitions computation between device and edge for purpose of coordinating the powerful cloud resource and the proximal edge resource for real-time DNN inference; (2) DNN right-sizing that further reduces computing latency via early exiting inference at an appropriate intermediate DNN layer. In addition, considering the potential network fluctuation in real-world deployment,Edgentis properly design to specialize for both static and dynamic network environment. Specifically, in a static environment where the bandwidth changes slowly,Edgentderives the best configurations with the assist of regression-based prediction models, while in a dynamic environment where the bandwidth varies dramatically,Edgentgenerates the best execution plan through the online change point detection algorithm that maps the current bandwidth state to the optimal configuration. We implementEdgentprototype based on the Raspberry Pi and the desktop PC and the extensive experimental evaluations demonstrateEdgent’s effectiveness in enabling on-demand low-latency edge intelligence.

Federated Learning (FL) has been proposed as an appealing approach to handle data privacy issue of mobile devices compared to conventional machine learning at the remote cloud with raw user data uploading. By leveraging edge servers as intermediaries to perform partial model aggregation in proximity and relieve core network transmission overhead, it enables great potentials in low-latency and energy-efficient FL. Hence we introduce a novel Hierarchical Federated Edge Learning (HFEL) framework in which model aggregation is partially migrated to edge servers from the cloud. We further formulate a joint computation and communication resource allocation and edge association problem for device users under HFEL framework to achieve global cost minimization. To solve the problem, we propose an efficient resource scheduling algorithm in the HFEL framework. It can be decomposed into two subproblems: resource allocation given a scheduled set of devices for each edge server and edge association of device users across all the edge servers. With the optimal policy of the convex resource allocation subproblem for a set of devices under a single edge server, an efficient edge association strategy can be achieved through iterative global cost reduction adjustment process, which is shown to converge to a stable system point. Extensive performance evaluations demonstrate that our HFEL framework outperforms the proposed benchmarks in global cost saving and achieves better training performance compared to conventional federated learning.

Fog computing is envisioned as a promising approach for supporting emerging computation-intensive applications on capacity and battery constrained mobile Internet of Things (IoT) devices. Technically speaking, a massive crowd of devices in close proximity can be harvested and collaborate for computation and communication resource sharing. Hence fog computing enables significant potentials in low-latency and energy-efficient mobile task execution. However, without an efficient incentive mechanism to stimulate resource sharing among devices, the benefits of fog computing cannot be fully realized. Leveraging coalitional game theory, this work presents an efficient incentive mechanism to incentivize mutually-beneficial resource cooperation among the devices for collaborative task execution. In particular, to efficiently achieve mutually beneficial task execution, the proposed mechanism groups the devices into multiple micro computing clusters (MCCs). Within each MCC, devices can exchange mutually beneficial actions by helping to compute or transmit tasks, making all of their performances no worse than local execution or execution in the fog server. The solution to the MCC formation is devised by both centralized and decentralized schemes and further proven to admit nice properties such as top coalition, core solution, individual rationality and computational efficiency. Extensive numerical studies demonstrate the superior performance of our MCC formation mechanisms.

Mobile edge computing (MEC) is emerging to support delay-sensitive 5G applications at the edge of mobile networks. When a user moves erratically among multiple MEC nodes, the challenge of how to dynamically migrate its service to maintain service performance (i.e., user-perceived latency) arises. However, frequent service migration can significantly increase operational cost, incurring the conflict between improving performance and reducing cost. To address these mis-aligned objectives, this paper studies the performance optimization of mobile edge service placement under the constraint of long-term cost budget. It is challenging because the budget involves the future uncertain information (e.g., user mobility). To overcome this difficulty, we devote to leveraging the power of prediction and advocate predictive service placement with predicted near-future information. By using two-timescale Lyapunov optimization method, we propose a T-slot predictive service placement (PSP) algorithm to incorporate the prediction of user mobility based on a frame-based design. We characterize the performance bounds of PSP in terms of cost-delay trade-off theoretically. Furthermore, we propose a new weight adjustment scheme for the queue in each frame named PSP-WU to exploit the historical queue information, which greatly reduces the length of queue while improving the quality of user-perceived latency. Rigorous theoretical analysis and extensive evaluations using realistic data traces demonstrate the superior performance of the proposed predictive schemes.

With the rapid development of mobile cellular technologies and the popularity of mobile devices, timely mobile traffic forecasting with high accuracy becomes more and more critical for proactive network service provisioning and efficient network resource allocation. Due to the complicated dynamic nature of mobile traffic demand, traditional time series methods cannot satisfy the requirements of prediction tasks well and often neglect the important spatial factors. In addition, while some recent approaches model mobile traffic prediction problem using temporal and spatial features, they only consider local geographical dependency and do not take influential distant regions into consideration. In this paper, we propose Graph Attention Spatial-Temporal Network (GASTN), a novel deep learning framework to tackle the mobile traffic forecasting problem. Specifically, GASTN considers spatial correlation through the geographical relation graph and utilizes structural recurrent neural networks to model the global near-far spatial relationships as well as capture the temporal dependencies between future demand for mobile traffic and historical traffic volume. Besides, two attention mechanisms are proposed to integrate different effects in a holistic way. Extensive experiments on a large-scale real-world mobile traffic dataset demonstrate that our model significantly outperforms the state-of-the-art methods.

Fog/edge computing has been recently regarded as a promising approach for supporting emerging mission-critical Internet of Things (IoT) applications on capacity and battery constrained devices. By harvesting and collaborating a massive crowd of devices in close proximity for computation, communication and caching resource sharing (i.e., 3C resources), it enables great potentials in low-latency and energy-efficient IoT task execution. To efficiently exploit 3C resources of fog devices in proximity, we propose F3C, a fog-enabled 3C resource sharing framework for energy-efficient IoT data stream processing by solving an energy cost minimization problem under 3C constraints. Nevertheless, the minimization problem proves to be NP-hard via reduction to a Generalized Assignment Problem (GAP). To cope with such challenge, we propose an efficient F3C algorithm based on an iterative task team formation mechanism which regards each task's 3C resource sharing as a subproblem solved by the elaborated min cost flow transformation. Via utility improving iterations, the proposed F3C algorithm is shown to converge to a stable system point. Extensive performance evaluations demonstrate that our F3C algorithm can achieve superior performance in energy saving compared to various benchmarks.

Mobile Edge Computing (MEC), envisioned as a cloud extension, pushes cloud resource from the network core to the network edge, thereby meeting the stringent service requirements of many emerging computation-intensive mobile applications. Many existing works have focused on studying the system-wide MEC service placement issues, personalized service performance optimization yet receives much less attention. Thus, in this paper we propose a novel adaptive user-managed service placement mechanism, which jointly optimizes a user's perceived-latency and service migration cost, weighted by user preferences. To overcome the unavailability of future information and unknown system dynamics, we formulate the dynamic service placement problem as a contextual Multi-armed Bandit (MAB) problem, and then propose a Thompson-sampling based online learning algorithm to explore the dynamic MEC environment, which further assists the user to make adaptive service placement decisions. Rigorous theoretical analysis and extensive evaluations demonstrate the superior performance of the proposed adaptive user-managed service placement mechanism.

Operating distributed Scrubbing Centers (SCs) to mitigate massive Distributed Denial of Service (DDoS) traffic in large-scale networks faces critical challenges. The operator needs to determine the diversion rule installation and elimination in the networks, as well as the scrubbing resource activation and revocation in the SCs, while minimizing the long-term cost and the cumulative decision-switching penalty without knowing the exact amount of the malicious traffic. We model and formulate this problem as an online nonlinear integer program. In contrast to many other online problems where future inputs are unknown but at least current inputs are known, a key new challenge here is that even part of the current inputs are unknown when decisions are made. To "learn" the best decisions online, we transform our problem via a gap-preserving approximation into an online optimization problem with only the known inputs, which is further relaxed and decoupled into a series of one-shot convex programs solvable in individual time slots. To overcome the intractability, we design a progressive rounding algorithm to convert fractional decisions into integral ones without violating the constraints. We characterize the competitive ratio of our approach as a function of the key parameters of our problem. We conduct evaluations using real-world data and confirm our algorithms' superiority over de facto practices and state-of-the-art methods.

Edge computing is envisioned to be the de-facto paradigm of hosting emerging low latency Internet-of-Things (IoT) data streaming services.For IoT data streaming in edge computing, cost management is of strategic significance, due to the low cost-efficiency of edge servers. While existing literature adopts a reactive approach to dynamically provisioning edge servers to reduce cost, the delay of server activation and instantiation has been mostly ignored. In this paper, we target a proactive approach to dynamic edge server provisioning for real-time IoT data streaming across edge nodes, which adjusts server provisioning ahead of time, based on prediction of the upcoming workload. To effectively predict upcoming workload, a learning-based method online gradient descent is applied. We further combine the online learning method with an online optimization algorithm for server provisioning in a joint online optimization framework, through (1) minimizing of the regret incurred by inaccurate workload prediction, and (2) minimizing the cost incurred by near-optimal online decisions. The resulting predictive online algorithm can well leverage the power of prediction and achieve a good performance guarantee, as verified by both rigorous theoretical analysis and extensive trace-driven evaluations.

To meet the stringent requirement of artificial intelligence applications, such as face recognition and video streaming analytics, a resource-constrained device can offload its task to nearby resource-rich devices in edge computing. Resource awareness, as a prime prerequisite for offloading decision-making, is critical for achieving efficient collaborative computation performance. In this paper, we consider cost-aware edge resource probing (CERP) framework design for infrastructure-free edge computing wherein a task device self-organizes its resource probing for informed computation offloading. We first propose a multi-stage optimal stopping formulation for the problem, and derive the optimal probing strategy which reveals a nice multi-threshold structure. Accordingly, we then devise a data-driven layered learning mechanism for more practical and complicated application environments. Layered learning enables the task device to adaptively learn the optimal probing sequence and decision thresholds at runtime, aiming at deriving a good balance between the gain of choosing the best edge device and the accumulated cost of deep resource probing. We further conduct thorough performance evaluation of the proposed CERP schemes using both extensive numerical simulations and realistic system prototype implementation, which demonstrate the superior performance of CERP in the diverse application scenarios.

Recently, the explosion of resource-hungry and delay-sensitive Internet-of-Things (IoT) applications as exemplified by wearable appliances, video surveillance, and connected vehicles have posed great challenges on the underlying IoT devices which typically have limited computation resource. In response, computation offloading is envisioned as a promising approach to augmenting capability of IoT devices. Toward real-time and efficient computation offloading, in this paper we propose a novel edge resource pooling framework, in which a massive crowd of devices at the network edge exploit device-to-device (D2D) collaboration for pooling and sharing computation resource with each other. Specifically, we first formulate the utility maximization problem under both computation and communication constraints as a mixed-integer linear programming problem, which is further proven to be NP-hard. To address this challenge, we propose a greedy heuristic based on the classical maximum network flow problem, and thus to schedule the task offloading in a cost-efficient manner. By considering the case that a centralized controller (e.g., a network operator) is not available, a decentralized task offloading scheme is further proposed, in which IoT devices communicate and determine D2D offloading strategy locally. Rigorous theoretical analysis and extensive evaluations demonstrate the effectiveness of the proposed algorithms.

With the rapid development of the Internet of Things (IoT) in the era of big data, the amount of collected data has increased dramatically. Data are one of the most important commodities in IoT. To maximize the utility of the collected data, it is crucial to design an open IoT data market that enables data owners and consumers to carry out data trading securely and efficiently. To address the challenge of security presented by an untrusted and nontransparent data market, we propose an edge/cloud-computing-assisted, blockchain-enhanced data market framework to support secure and efficient IoT data trading, with a particular focus on an optimal pricing mechanism. In this mechanism, an authorized market-agency works as a scheduler, determining the win-owner and its pricing strategy to the consumer. We formulate a two-stage Stackelberg game to solve the pricing and purchasing problem of the data consumer and the market-agency. In the first stage of the game, the market-agency gives the win-owner and its pricing strategy. In the second stage, the data consumer decides on its purchasing quantity of data. We consider competition between data owners and propose a competition-enhanced pricing scheme (CPS). We apply backward induction to analyze the subgame perfect equilibrium at each stage for both independent and CPSs. Lastly, we validate the existence and uniqueness of Stackelberg equilibrium, and the numerical results show the efficiency of the CPS.

In this paper, we propose Chimera, a novel hybrid edge computing framework, integrated with the emerging edge cloud radio access network, to augment network-wide vehicle resources for future large-scale vehicular crowdsensing applications, by leveraging a multitude of cooperative vehicles and the virtual machine (VM) pool in the edge cloud via the control of the application manager deployed in the edge cloud. We present a comprehensive framework model and formulate a novel multivehicle and multitask offloading problem, aiming at minimizing the energy consumption of network-wide recruited vehicles serving heterogeneous crowdsensing applications, and meanwhile reconciling both application deadline and vehicle incentive. We invoke Lyapunov optimization framework to design TaskSche, an online task scheduling algorithm, which only utilizes the current system information. As the core components of the algorithm, we propose a task workload assignment policy based on graph transformation and a knapsack-based VM pool resource allocation policy. Rigorous theoretical analyses and extensive trace-driven simulations indicate that our framework achieves superior performance (e.g., 20%-68% energy saving without overstepping application deadlines for network-wide vehicles compared with vehicle local processing) and scales well for a large number of vehicles and applications.

Recently, location prediction has attracted considerable research effort because of the popularity of location-based services, such as mobile advertising and recommendations. With the unprecedented proliferation of mobile social networks, such as WeChat and Twitter, we are able to use location service to bridge the online and offline worlds, which is of great significance to many smart city applications. Different from existing studies, in this paper, we promote a user-centric location prediction approach by leveraging a user's local mobile social information without involving other users' location privacy. We propose a factor graph learning model that integrates not only user's social and network information but also the correlations between a user's locations into a unified framework. Furthermore, we use ReliefF algorithm to select user-specific significant features for location prediction and define the measure of location entropy to study the similarity between location, network status, and social behavior. To show the benefit of precise location prediction, we further apply it to personalized service migration in mobile edge computing (MEC) and accordingly propose prediction-based amortizing algorithm and lazy migration algorithm that can well balance the tradeoff between migration cost and non-migration latency in a cost-efficient manner. We conduct extensive experiments using a real-world data trace, which shows that our model performs much better in location prediction compared with several classic methods and the MEC service quality can be significantly enhanced by leveraging the location prediction.

Mobile edge computing (MEC) offers a way to shorten the cloud servicing delay by building the small-scale cloud infrastructures, such as cloudlets at the network edge, which are in close proximity to end users. On one hand, it is energy consuming and costly to place each cloudlet on each access point (AP) to process the requested tasks. On the other hand, the service provider should provide delay-guaranteed service to end users, otherwise they may get revenue loss. In this paper, we first model how to calculate the task completion delay in MEC and mathematically analyze the energy consumption of different equipments in MEC. Subsequently, we study how to place cloudlets on the network and allocate each requested task to cloudlets and public cloud with the minimum total energy consumption without violating each task's delay requirement. We prove that this problem is NP-hard and propose a Benders decomposition-based algorithm to solve it. We also present a software-defined network (SDN)-based framework to deploy the proposed algorithm. Extensive simulations reveal that the proposed algorithm can achieve an (close-to-)optimal performance in terms of energy consumption and acceptance ratio compared with two benchmark heuristics.

The blockchain-empowered mobile-edge computing (MEC) is a promising solution for enhancing the computation capabilities of mobile equipments (MEs) to process computation-intensive tasks such as the real-time data processing tasks and mining tasks. However, because of the “cold start” and “long return” problems, efficient computation offloading cannot be achieved in blockchain-empowered MEC because the MEs do not always have enough coins to afford the offloading service cost. In this article, we study the joint computation-offloading and coin-loaning problem for blockchain-empowered MEC to minimize the total cost of all MEs. We introduce the banks that can provide loan services to the MEs to address the above two issues. We formulate the problem as a noncooperative game to model the competitions between the myopic MEs. By using a potential game method, we prove the existence of a pure-strategy Nash equilibrium (NE) and design a distributed algorithm to achieve the NE point with low computational complexity. We also provide an upper bound on the price of anarchy of the game by theoretical proof. Besides, two smart contracts are designed to automatically perform the computing resource trading and coin loaning processes. Lastly, our simulation results show that our proposed algorithm can significantly reduce the total cost of all MEs, has better performance compared with other solutions, and scales well as the number of MEs increases. Moreover, the financial cost for executing the two smart contracts on the Ethereum network is low.

Edge computing (EC) has quickly ascended to be the de-facto standard for hosting emerging low-latency applications, as exemplified by intelligent video surveillance, Internet of Vehicles, and augmented reality. For EC, service function chaining is envisioned as a promising approach to configure various services in an agile, flexible, and cost-efficient manner. When running on top of geographically dispersed edge clouds, fully unleashing the benefits of service function chaining is, however, by no means trivial. In this paper, we propose an online orchestration framework for cross-edge service function chaining, which aims to maximize the holistic cost efficiency, via jointly optimizing the resource provisioning and traffic routing on-the-fly. This long-term cost minimization problem is difficult since it is NP-hard and involves future uncertain information. To simultaneously address these dual challenges, we carefully combine an online optimization technique with an approximate optimization method in a joint optimization framework, through: 1) decomposing the long-term problem into a series of one-shot fractional problem with a regularization technique and 2) rounding the fractional solution to a near-optimal integral solution with a randomized dependent scheme that preserves the solution feasibility. The resulting online algorithm achieves an outstanding performance guarantee, as verified by both rigorous theoretical analysis and extensive trace-driven simulations.

With the breakthroughs in deep learning, the recent years have witnessed a booming of artificial intelligence (AI) applications and services, spanning from personal assistant to recommendation systems to video/audio surveillance. More recently, with the proliferation of mobile computing and Internet of Things (IoT), billions of mobile and IoT devices are connected to the Internet, generating zillions bytes of data at the network edge. Driving by this trend, there is an urgent need to push the AI frontiers to the network edge so as to fully unleash the potential of the edge big data. To meet this demand, edge computing, an emerging paradigm that pushes computing tasks and services from the network core to the network edge, has been widely recognized as a promising solution. The resulted new interdiscipline, edge AI or edge intelligence (EI), is beginning to receive a tremendous amount of interest. However, research on EI is still in its infancy stage, and a dedicated venue for exchanging the recent advances of EI is highly desired by both the computer system and AI communities. To this end, we conduct a comprehensive survey of the recent research efforts on EI. Specifically, we first review the background and motivation for AI running at the network edge. We then provide an overview of the overarching architectures, frameworks, and emerging key technologies for deep learning model toward training/inference at the network edge. Finally, we discuss future research opportunities on EI. We believe that this survey will elicit escalating attentions, stimulate fruitful discussions, and inspire further research ideas on EI.

While modern wireless devices are capable of using multiple WiFi interfaces, the Multipath TCP (MPTCP) protocol has been employed to make full use of the capacity of many radios by enabling multiple path communication simultaneously. To provide exceptional mobility support in wireless networks, a key question is to determine the best handoff strategy to switch between access points or among WiFi/3G interfaces during roaming. In this paper, we propose SERO, a novel model-driven SEamless ROaming framework to optimize layer-2 handoff and vertical handoff for multihomed devices using MPTCP. The proposed framework adopts a measurement-based method to derive the TCP throughput model for wireless communication during handoff. Based on the throughput model, we propose a hybrid handoff strategy that uses multiple WiFi interfaces for data transmission and employs 3G augmentation to bridge the network interruption caused by handoff and to guarantee the total throughput above a predefined threshold for roaming devices. We implement the SERO framework in a real-deployed wireless mesh network testbed, and evaluate its performance by extensive experiments, which shows that SERO achieves performance gain of 26%-180% compared with several existing handoff strategies.

Edge computing emerges as a promising paradigm to decentralize computation power to the edge of the network and thus improve user experience by task offloading. A user can perfectly schedule his tasks to be executed on edge servers if the execution time of all tasks can be known beforehand. However, it is difficult to know the task execution time (TET) before performing actual offloading, which normally varies on edge servers with different software and hardware configurations. Moreover, such configuration information is not always available to end users due to security concerns. In this paper, we first propose a learning-driven algorithm to accurately predict TETs of all tasks in such an asymmetrically informed edge computing environment. The basic idea is to predict unknown TETs using only a small sampled set of TETs by exploiting the underlying correlation between TETs and edge server configurations. Next, we formulate the problem of task offloading into a constrained optimization problem, which is unfortunately proved to be NP-hard. To address the above challenge, we design a task offloading algorithm, called Maximum Efficiency First Ordered (MEFO), to achieve near-optimal efficiency. Field measurements and experiments have been conducted to demonstrate that our proposed learning-driven algorithm can predict TETs more accurately than other algorithms as long as the fraction of sampled TETs is larger than a small predefined threshold, and our proposed MEFO algorithm achieves a much higher success rate of task offloading and a shorter processing delay with very limited information of edge servers.

Recently, location prediction has attracted considerable research effort because of the popularity of location- based services, such as mobile advertising and recommendations. With the unprecedented proliferation of mobile social networks, we are able to use location service to bridge the online and offline worlds. Different from existing studies, in this paper we promote a user-centric location prediction approach by leveraging a user's local mobile social information without involving other users' location privacy. We propose a factor graph learning model that integrates not only user's social and network information, but also the correlations between user's locations into a unified framework. Furthermore, we use ReliefF algorithm to select user-specific significant features for location prediction and define the measure of location entropy to study the similarity between location, network status and social behavior. We conduct extensive experiments using a real-world dataset, which shows that our model performs much better in location prediction compared with several classic methods.

With the pervasive availability of LCD displays and phone cameras, screen-camera communication has attracted grate attentions due to the characteristics of convenience, security, infrastructure-free, and contactless. The existing screen-camera communication systems using dynamic barcodes suffer from poor ability of color recognition. In this paper, we propose a machine learning based multi-color dynamic barcode system called MMCode to overcome such limit. We formulate the color recognition problem as a machine learning task, and propose a semi-supervised clustering algorithm to achieve finer-grained color recognition. The proposed mechanism inserts reference colors in the barcode design, and adopts a semi-supervised Gaussian Mixed Model (GMM) algorithm for frame decoding. We implement MMCode as an Android APP and test its performance under real screen-camera communication scenarios. Extensive experiments show that the proposed MMCode achieves significant enhancement on the capacity of dynamic barcodes compared to the state-of-the-arts.

Named Data Networking (NDN) is a promising technology for content centric networks, and it is suitable for vehicular networks since no IP architecture is required. Quite a number of solutions have been proposed for vehicular NDN (V- NDN), but high communication cost due to frequent topology changes caused by high mobility of vehicles is still a challenge to be addressed. In this paper, we study how to disseminate traffic information to vehicles via V-NDN. Different from existing works, we consider navigation route based data interests, i.e., a vehicle is concerned about the traffic information along road segments planned to take. According to such a data interest scenario, we propose a tree based data interest structure and associated maintenance operations to merge identical data interests due to overlapping navigation routes among different vehicles. With the tree based data interest management, the number of interest packets can be significantly reduced. Then, we propose trigger based mechanisms for data interest packet re-sending and forwarding, which can avoid unnecessary interest packets re-sending. With our design, traffic information can be disseminated to interested nodes with high success ratio and low communication cost simultaneously. Simulations via SUMO and ndnSIM confirm such advantages of our work.

Mobile edge computing is a new computing paradigm in which cloud computing capabilities are pushed from the network core to the network edge to serve the end-user in proximity. However, with the sinking of computing capabilities, the new challenge incurred by user mobility arises: since end-users typically move erratically, the services should be dynamically migrated among multiple edges to maintain the service performance, i.e., user-perceived latency. Tackling this problem is non-trivial since frequent service migration would greatly increase the operational cost. To address this challenge in terms performance-cost trade-off, in this paper we study the mobile edge service performance optimization problem under long-term cost budget constraint. To address user mobility which is typically unpredictable, we first apply Lyapunov optimization to decompose the long-term optimization problem into a series of real-time optimization problems which do not require a priori knowledge such as user mobility. As the decomposed problem is NP-hard, we further propose an efficient heuristic based on the Markov approximation technique. Rigorous theoretical analysis and extensive evaluations demonstrate the efficacy of the proposed solution.

We study a privacy-preserving data collection problem, by jointly considering data reporters' data correlation and social relationship. A data collector gathers data from individuals to perform a certain analysis with a privacy-preserving mechanism. Due to data correlation, the data analysis based on the reported data can cause privacy leakage to other individuals (even if they do not report data). The data reporters will take such a privacy threat into account, owing to the social relationship among individuals. This motivates us to formulate a two-stage Stackelberg game: In Stage I, the data collector selects some individuals as data reporters and designs a privacy-preserving mechanism for a sum query analysis. In Stage II, the selected data reporters contribute their data with possible perturbations (through adding noise). By analyzing the data reporters' equilibrium decisions in Stage II, we show that given any fixed reporter set, only one data reporter with the most significant joint consideration of the social relationship and data correlation may add noise to his reported data. The rest of the data reporters will truthfully report their data. In Stage I, we derive the data collector's optimal privacy-preserving mechanism and propose an efficient algorithm to select the data reporters. We conclude that the data collector should jointly capture the impact of data correlation and social relation to ensure all data reporters truthfully reporting their data. We conduct extensive simulations based on random network and real-world social data to investigate the impact of data correlation and social network on the system. We find that the availability of social network information is more critical to the data collector compared with data correlation information.

Software Defined Networks (SDN) have been proposed for use in applications of the Internet of Things (IoT), termed as software-defined IoT (SD-IoT) network, because of the popularity and capability of mobile devices being used for networking in relatively large areas. However, a single controller in SDN has a limited request-processing capability, so a distributed control plane with multiple physical controllers has been used to achieve scalability and reliability for supporting the IoT applications. Accordingly, the data plane of an SDN is partitioned into multiple domains, and each controller just takes over one. When considering both delays and loads of requests to the controllers, a partitioning problem arises. It is required to consider the distributions of flow paths, since inter-domain flow paths will create an extra load of requests to the controllers. In this paper, we investigate the partitioning problem in SD-IoT networks. Since uploading sensing data through the IoT gateways are non-uniform, we utilize a hypergraph to model the relationship between the spatial events and the gateways in IoT for urban sensing. We propose a Partitioning Algorithm for Software-defined IoT Network (PASIN) to partition the SDN by considering both delays and loads of requests from the flow paths. Our extensional simulations verify the effectiveness of our proposed approach.

The explosion of resource-hungry mobile applications has posed great challenges on the underlying mobile devices which typically have limited computation resource. In response, device-to-device (D2D) computation offloading is envisioned as a promising approach to the problem by gearing resource-rich devices and resource-poor devices. Towards real-time and efficient computation offloading, in this paper, we proposed a novel edge resource pooling framework called ERP, in which a massive crowd of devices at the network edge exploit D2D collaboration for pooling and sharing computation resource with each other. Specifically, we first formulate the utility maximization problem under both computation and communication constraints as a mixed-integer linear programming (MILP), which is further proven to be NP-hard. To address this challenge, we propose a centralized greedy heuristic based on the classical maximum network flow problem, which schedules the task offloading in a cost-efficient manner. Rigorous theoretical analysis and extensive evaluations demonstrate the effectiveness of the heuristic to some extent.

A homogeneous fog network is defined as a group of peer nodes with sharable computing and storage resources, as well as spare spectrum for node-to-node/device-to-device communications and task scheduling. It promotes more intelligent applications and services in different Internet of Things (IoT) scenarios, thanks to effective collaborations among neighboring fog nodes via cognitive spectrum access techniques. In this paper, a comprehensive analytical model that considers circuit, computation, offloading energy consumptions is developed for accurately evaluating the overall energy efficiency (EE) in homogeneous fog networks. With this model, the tradeoff relationship between performance gains and energy costs in collaborative task offloading is investigated, thus enabling us to formulate the EE optimization problem for future intelligent IoT applications with practical constraints in available computing resources at helper nodes and unused spectrum in neighboring environments. Based on rigorous mathematical analysis, a maximal energy-efficient task scheduling (MEETS) algorithm is proposed to derive the optimal scheduling decision for a task node and multiple neighboring helper nodes under feasible modulation schemes and time allocations. Extensive simulation results demonstrate the tradeoff relationship between EE and task scheduling performance in homogeneous fog networks. Compared with traditional task scheduling strategies, the proposed MEETS algorithm can achieve much better EE performance under different network parameters and service conditions.

In a device-to-device (D2D) communications underlaying cellular network, any user is a potential eavesdropper for the transmissions of others that occupy the same spectrum. The physical-layer security mechanism of theoretical secure capacity, which maximizes the rate of reliable communication from the source user to the legitimate receiver and ensure unauthorized users learn as little as information as possible, is typically employed to guarantee secure communications. As hand-held devices are carried by human beings, we may leverage their social trust to decrease the number of potential eavesdroppers. Aiming to establish a new paradigm for solving the challenging problem of security and efficiency tradeoff, we propose a social trust-aware D2D communication architecture that exploits the social-domain trust for securing the physical-domain communication. In order to understand the impact of social trust on the security of transmissions, we analyze the system ergodic rate of social trust aided communications via stochastic geometry, and our result based on a real data set shows that the proposed social trust aided D2D communication increases the system secrecy rate by about 63% compared with the scheme without considering social trust relation. Furthermore, in order to provide implementation mechanism, we utilize matching theory to implement efficient resource allocation among multiple users. Numerical results show that our proposed mechanism increases the system secrecy rate by 28% with fast convergence over the social oblivious approach.

Mobile edge computing is a new computing paradigm, which pushes cloud computing capabilities away from the centralized cloud to the network edge. However, with the sinking of computing capabilities, the new challenge incurred by user mobility arises: since end users typically move erratically, the services should be dynamically migrated among multiple edges to maintain the service performance, i.e., user-perceived latency. Tackling this problem is non-trivial since frequent service migration would greatly increase the operational cost. To address this challenge in terms of the performance-cost tradeoff, in this paper, we study the mobile edge service performance optimization problem under long-term cost budget constraint. To address user mobility which is typically unpredictable, we apply Lyapunov optimization to decompose the long-term optimization problem into a series of real-time optimization problems which do not require a priori knowledge such as user mobility. As the decomposed problem is NP-hard, we first design an approximation algorithm based on Markov approximation to seek a near-optimal solution. To make our solution scalable and amenable to future fifth-generation application scenario with large-scale user devices, we further propose a distributed approximation scheme with greatly reduced time complexity, based on the technique of the best response update. Rigorous theoretical analysis and extensive evaluations demonstrate the efficacy of the proposed centralized and distributed schemes.

In this paper, we advocate edge caching in cloud radio access networks (C-RAN) to facilitate the ever-increasing mobile multimedia services. In our framework, central offices will cooperatively allocate cloud resources to cache popular contents and satisfy user requests for those contents, so as to minimize the system costs in terms of storage, VM reconfiguration, content access latency, and content migration. However, this joint resource allocation, content placement and request routing, is nontrivial, since it needs to be continuously adjusted to accommodate system dynamics, such as user movement and content slashdot effect, while taking into account the time-correlated adjustment costs for VM reconfiguration and content migration. To this end, we build a comprehensive model to capture the key components of edge caching in C-RAN and formulate a joint optimization problem, aiming at minimizing the system costs over time and meanwhile satisfying the time-varying user requests and respecting various practical constraints (e.g., storage and bandwidth). Then, we propose a novel online approximation algorithm by resorting to the regularization, rounding, and decomposition technique, which can be proved to have a parameterized competitive ratio with a polynomial running time. Extensive trace-driven simulations corroborate the efficiency, flexibility, and lightweight of our proposed online algorithm; for instance, it achieves an empirical competitive ratio around 2 - 4 and gains over 30% improvement compared with many state-of-the-art algorithms in various system settings.

We study a mechanism design problem of privacy- preserving data collection with privacy protection uncertainty. A data collector wants to collect enough data to perform a certain computation that benefits the individuals who contribute the data, with the possibility of individual privacy leakage. The data collector adopts a privacy-preserving mechanism by adding some random noise to the computation result, which reduces the accuracy of the computation. Individuals decide whether to contribute data based on the potential benefit and the possible privacy cost induced by the mechanism. Due to the intrinsic uncertainty involved in privacy protection, we model individuals' privacy-aware participation using the prospect theory, which more accurately models individuals' behavior under uncertainty than the traditional expected utility theory. We show that the data collector's utility maximization problem involves a polynomial of high and fractional order, which is difficult to solve analytically. We get around this issue by proposing an approximation method, which allows us to obtain a closed form unique solution of the data collector's decision problem. We numerically show that the approximation error is small when the number of individuals is large. By comparing with the results under the expected utility theory, we conclude that a data collector who considers the more realistic prospect theory modeling should adopt a stricter privacy-preserving mechanism to boost her utility.

In this paper we propose HyFog, a novel hybrid task offloading framework in fog computing, where device users have the flexibility of choosing among multiple options for task executions, including local mobile execution, Device-to-Device (D2D) offloaded execution, and Cloud offloaded execution. We further develop a novel three-layer graph matching algorithm for efficient hybrid task offloading among the devices. Specifically, we first construct a three-layer graph to capture the choice space enabled by these three execution approaches, and then the problem of minimizing the total task execution cost is recast as a minimum weight matching problem over the constructed three-layer graph, which can be efficiently solved using the Edmonds's Blossom algorithm. Numerical results demonstrate that the proposed three-layer graph matching solution can achieve superior performance, with more than 50% cost reduction over the case of local task executions by all the devices.

The ever increasing of display resolution on mobile devices raises high demand for GPU rendering details. However, the challenge of poor hardware support but fine-grained rendering details often makes user unsatisfied especially in calling for high frame rate scenarios, e.g., game. To resolve such issue, we propose BUTTERFLY, a novel system which collaboratively utilizes mobile GPUs to process high-quality rendering details for on-the-go mobile users. In particular, ButterFly achieves two technical contributions for the collaborative design: (1) a mobile device can migrate GPU workloads in buffer queue to peers, and (2) the collaborative rendering mechanism benefits user high quality details while significant power saving performance. Both techniques are compatible with the OpenGL ES standards. Furthermore, a 40-person survey perceives that ButterFly can provide excellent user experience of both rendering details and frame rate over Wi-Fi network. In addition, our comprehensive trace-driven experiments on Android prototype reveal the benefits of Butterfly have more superior performance over state-of-the-art systems, which achieves more than 28.3% power saving.

The rapid growth of online social networks has strengthened wireless users' social relationships, which in turn has resulted in more data traffic due to network effect in the social domain. Nevertheless, the boosted demand for wireless services may challenge the limited wireless capacity. To build a thorough understanding, we study mobile users' data usage behavior by jointly considering the network effect due to their social relationships in the social domain and the congestion effect in the physical wireless domain. Specifically, we develop a Stackelberg game for socially aware data usage: in Stage I, a wireless provider first decides the data pricing to all users in order to maximize its revenue, and then in Stage II, users decide their data usage, for the given price, subject to mutual interactions under both social network effect and congestion effect. We analyze the two-stage game via backward induction. In particular, for Stage II, we first provide conditions for the existence and the uniqueness of a user demand equilibrium (UDE). Then, we propose algorithms to find the UDE and for users to reach the UDE in a distributed manner. We further investigate the impact of different system parameters on the UDE. Next, for Stage I, we develop an optimal pricing algorithm to maximize the wireless provider's revenue. We numerically evaluate the performance of our proposed algorithms using real data, and thereby draw useful engineering insights for the operation of wireless providers: 1) when social network effect dominates congestion effect, the marginal gain of the total usage increases with the social ties and the number of users, or decreases with the congestion coefficient; in contrast, when congestion effect dominates social network effect, the marginal gain decreases (or increases, respectively) with these parameters and 2) when social network effect is strong, a lower price should be set to increase the total revenue; in contrast, when congestion effect is strong, a higher price is preferred.

Recent years have witnessed the proliferation of mobile crowdsourcing that brings a new opportunity to leverage human intelligence and movement behaviors to wider application areas. In parallel with the development of online centralized platforms, we look into the realization of self-organized mobile crowdsourcing drawing on opportunistic networks, and propose the Crowd Foraging framework, in which a mobile task requester can proactively recruit a massive crowd of opportunistic encountered mobile workers in real time for quick and high-quality results. We present a comprehensive framework model that fully integrates human behavior factors for modeling task profile, worker arrival, and work ability, and then introduce a service quality concept to indicate the expected service gain that a requester can enjoy when she recruits an arrival worker by jointly considering the work ability of workers as well as timeliness and reward of tasks. Furthermore, we formulate a sequential worker recruitment problem as an online multiple stopping problem to maximize the expected sum of service quality, and accordingly derive an optimal worker recruitment policy through the dynamic programming principle, which exhibits a nice threshold-based structure. We provide data-driven case studies to validate the assumptions used in the policy design, and conduct extensive trace-driven numerical evaluations, which demonstrate that our policy can achieve superior performance (e.g., improve more than 30% performance over classic policies). Besides, our Android prototype shows that the Crowd Foraging framework is cost-efficient, such as requiring less than 7 s and 6 J in terms of time and energy consumption for the optimal threshold calculation in our policy in most cases.

Distributed spectrum access (DSA) is challenging, since an individual secondary user often has limited sensing capabilities only. One key insight is that channel recommendation among secondary users can help to take advantage of the inherent correlation structure of spectrum availability in both time and space, and enable users to obtain more informed spectrum opportunities. With this insight, we advocate to leverage the wisdom of crowds, and devise social recommendation aided DSA mechanisms to orient secondary users to make more intelligent spectrum access decisions, for both strong and weak network information cases. We start with the strong network information case where secondary users have the statistical information. To mitigate the difficulty due to the curse of dimensionality in the stochastic game approach, we take the one-step Nash approach and cast the social recommendation aided DSA decision making problem at each time slot as a strategic game. We show that it is a potential game, and then devise an algorithm to achieve the Nash equilibrium by exploiting its finite improvement property. For the weak information case where secondary users do not have the statistical information, we develop a distributed reinforcement learning mechanism for social recommendation aided DSA based on the local observations of secondary users only. Appealing to the maximum-norm contraction mapping, we also derive the conditions under which the distributed mechanism converges and characterize the equilibrium therein. Numerical results reveal that the proposed social recommendation aided DSA mechanisms can achieve a superior performance using real social data traces and its performance loss in the weak network information case is insignificant, compared with the strong network information case.

With increasing popularity of location-based services (LBSs), there have also been growing concerns for location privacy. To protect location privacy in an LBS, mobile users in physical proximity can work in concert to collectively change their pseudonyms, in order to hide spatial-temporal correlation in their location traces. In this paper, we leverage mobile users' social tie structure to motivate them to participate in pseudonym change. Drawing on a social group utility maximization framework, we cast users' decision making of whether to change pseudonyms as a socially aware pseudonym change game (SA-PCG). The SA-PCG further assumes a general anonymity model that allows a user to have its specific anonymity set for personalized location privacy. For the SA-PCG, we show that there exists a socially aware Nash equilibrium (SNE), and quantify the system efficiency of SNEs with respect to the optimal social welfare. Then, we develop a greedy algorithm that myopically determines users' strategies, based on the social group utility derived from only the users whose strategies have already been determined. We show that this algorithm efficiently finds an SNE that enjoys desirable properties: 1) it is socially aware coalition-proof, and thus is also Pareto-optimal; 2) it achieves higher social welfare than any SNE for the socially oblivious pseudonym change game. We further quantify the system efficiency of this SNE with respect to the optimal social welfare. We also show that this SNE can be achieved in a distributed manner. Numerical results using real data corroborate that social welfare can be significantly improved by exploiting social ties.

Mobile online social networks (OSNs) are emerging as the popular mainstream platform for information and content sharing among people. In order to provide the quality of experience support for mobile OSN services, in this paper, we propose a socially-driven learning-based framework, namely Spice, for the media content prefetching to reduce the access delay and enhance mobile user's satisfaction. Through a large-scale data-driven analysis over real-life mobile Twitter traces from over 17 000 users during a period of five months, we reveal that the social friendship has a great impact on user's media content click behavior. To capture this effect, we conduct the social friendship clustering over the set of user's friends, and then develop a cluster-based Latent Bias Model for socially-driven learning-based prefetching prediction. We then propose a usage-adaptive prefetching scheduling scheme by taking into account that different users may possess heterogeneous patterns in the mobile OSN app usage. We comprehensively evaluate the performance of Spice framework using trace-driven emulations on smartphones. Evaluation results corroborate that the Spice can achieve superior performance, with an average 80.6% access delay reduction at the low cost of cellular data and energy consumption. Furthermore, by enabling users to offload their machine learning procedures to a cloud server, our design can achieve up to a factor of 1000 speed-up over the local data training execution on smartphones.

Cloudlet is a new paradigm in mobile cloud computing to provide resources to nearby mobile users via one-hop wireless connections. In this study, we leverage the social tie structure among mobile users to achieve mutual-beneficial computation offloading decision making and hence enhance the system-wide performance.Drawing on a social group utility maximization (SGUM) framework, we cast users' decision making of whether to offload or not as a social-aware computation offloading game (COG). We analyze the structural property of the SGUM-based COG (SCOG) and show that there exists a social-aware Nash equilibrium (SNE). We then design a distributed computation offloading algorithm that can achieve the SNE of SCOG and quantify its performance gap with respect to the social optimal solution. Numerical results show that the computation offloading performance can be significantly enhanced by leveraging the social ties among the users.

Computation offloading has already shown itself to be successful for enabling resource-intensive applications on mobile devices. However, in view of immersive applications, the offloaded tasks could be duplicate when multiple users are in the same environment. In this paper, we consider the scenario that multiple mobile users offload duplicated computation tasks to a set of nearby Femto-Cloud called Small Cell cloud enhanced e-NodeB (SCceNB), and share the computation results among them. Our goal is to find an optimal offloading and sharing strategy to minimize the overall energy consumption at the mobile terminal side. To this end, we propose a cooperative call graph to model the problem. Based on the derived call graph, we present a distributed algorithm that combines notions from 0-1 programming and coalitional game to solve it, while considering the delay constraint for each mobile user as well as the computation ability and memory constraints of each SCceNB. Simulation results show that our proposal can reduce energy consumption up to 39.73%, 34.37%, and 19.54% compared to the “total offloading” scheme, the “no offloading” scheme, and the “optimal offloading without sharing” scheme, respectively.

In this paper, we advocate Crowdlet, a novel self-organized mobile crowdsourcing paradigm, in which a mobile task requester can proactively exploit a massive crowd of encountered mobile workers at real-time for quick and high-quality results. We present a comprehensive system model of Crowdlet that defines task, worker arrival and worker ability models. Further, we introduce a service quality concept to indicate the expected service gain that a requester can enjoy when he recruits an encountered worker, by jointly taking into account worker ability, real-timeness and task reward. Based on the models, we formulate an online worker recruitment problem to maximize the expected sum of service quality. We derive an optimal worker recruitment policy through the dynamic programming principle, and show that it exhibits a nice threshold based structure. We conduct extensive performance evaluation based on real traces, and numerical results demonstrate that our policy can achieve superior performance and improve more than 30% performance gain over classic policies. Besides, our Android prototype shows that Crowdlet is cost-efficient, requiring less than 7 seconds and 6 Joule in terms of time and energy cost for policy computation in most cases.

Mobile online social networks (OSNs) are emerging as the popular mainstream platform for information and content sharing among people. In order to provide Quality of Experience (QoE) support for mobile OSN services, in this paper we propose a socially-driven learning-based framework, namely Spice, for media content prefetching to reduce the access delay and enhance mobile user's satisfaction. Through a large-scale data-driven analysis over real-life mobile Twitter traces from over 17,000 users during a period of five months, we reveal that the social friendship has a great impact on user's media content click behavior. To capture this effect, we conduct social friendship clustering over the set of user's friends, and then develop a cluster-based Latent Bias Model for socially-driven learning-based prefetching prediction. We then propose a usage-adaptive prefetching scheduling scheme by taking into account that different users may possess heterogeneous patterns in the mobile OSN app usage. We comprehensively evaluate the performance of Spice framework using trace-driven emulations on smartphones. Evaluation results corroborate that the Spice can achieve superior performance, with an average 67.2% access delay reduction at the low cost of cellular data and energy consumption. Furthermore, by enabling users to offload their machine learning procedures to a cloud server, our design can achieve speed-up of a factor of 1000 over the local data training execution on smartphones.

Exponential growth of media consumption in online social networks demands effective recommendation to improve the quality of experience especially for on-the-go mobile users. By means of large-scale trace-driven measurements over mobile Twitter traces from users, we reveal the significance of affective features in shaping users' social media behaviors. Existing recommender systems however, rarely support this psychological effect in real-life. To capture this effect, in this paper we propose Kaleido, a real mobile system to achieve an affect-aware learning-based social media recommendation.Specifically, we design a machine learning mechanism to infer the affective feature within media contents. Furthermore, a cluster-based latent bias model is provided for jointly training the affect, behavior and social contexts. Our comprehensive experiments on Android prototype expose a superior prediction accuracy of 82%, with more than 20% accuracy improvement over existing mobile recommender systems. Moreover, by enabling users to offload their machine learning procedures to the deployed edge-cloud testbed, our system achieves speed-up of a factor of 1,000 against the local data training execution on smartphones.

In this paper, we propose device-to-device (D2D) Fogging, a novel mobile task offloading framework based on network-assisted D2D collaboration, where mobile users can dynamically and beneficially share the computation and communication resources among each other via the control assistance by the network operators. The purpose of D2D Fogging is to achieve energy efficient task executions for network wide users. To this end, we propose an optimization problem formulation that aims at minimizing the time-average energy consumption for task executions of all users, meanwhile taking into account the incentive constraints of preventing the over-exploiting and free-riding behaviors which harm user's motivation for collaboration. To overcome the challenge that future system information such as user resource availability is difficult to predict, we develop an online task offloading algorithm, which leverages Lyapunov optimization methods and utilizes the current system information only. As the critical building block, we devise corresponding efficient task scheduling policies in terms of three kinds of system settings in a time frame. Extensive simulation results demonstrate that the proposed online algorithm not only achieves superior performance (e.g., it reduces approximately 30% ~ 40% energy consumption compared with user local execution), but also adapts to various situations in terms of task type, user amount, and task frequency.

To meet the explosive demand on delivering high-definition video steams over cellular networks, we design a Social-aware video multiCast (SoCast) system leveraging device-to-device (D2D) communications. One salient feature of SoCast is to stimulate effective cooperation among mobile users (clients), by making use of two types of important social ties, i.e., social trust and social reciprocity. By using SoCast, clients form groups to obtain missing packets from other clients and restore incomplete video frames, according to the unique video encoding structure. In return, the user perception of the mobile video quality can be substantially improved. Specifically, we first cast the problem of social ties based group formation among clients for cooperative video multicast as a coalitional game, and then devise a distributed algorithm to obtain the core solution (group formation) for the formulated coalitional game. Further, a resource allocation scheme is proposed for the base station to handle D2D radio resource requests from client groups. Extensive numerical studies using real video traces corroborate the significant gain using SoCast.

We develop a social group utility maximization (SGUM) framework for cooperative wireless networking that takes into account both social relationships and physical coupling among users. Specifically, instead of maximizing its individual utility or the overall network utility, each user aims to maximize its social group utility that hinges heavily on its social tie structure with other users. We show that this framework provides rich modeling flexibility and spans the continuum between non-cooperative game and network utility maximization (NUM)-two traditionally disjoint paradigms for network optimization. Based on this framework, we study three important applications of SGUM, in database assisted spectrum access, power control, and random access control, respectively. For the case of database assisted spectrum access, we show that the SGUM game is a potential game and always admits a socially-aware Nash equilibrium (SNE). We also develop a distributed spectrum access algorithm that can converge to the SNE and also quantify the trade-off between the performance and convergence time of the algorithm. For the cases of power control and random access control, we show that there exists a unique SNE and the network performance improves as the strength of social ties increase. Numerical results corroborate that the SGUM solutions can achieve superior performance using real social data trace. Furthermore, we show that the SGUM framework can be generalized to take into account both positive and negative social ties among users, which can be a useful tool for studying network security problems.

Mobile-edge cloud computing is a new paradigm to provide cloud computing capabilities at the edge of pervasive radio access networks in close proximity to mobile users. In this paper, we first study the multi-user computation offloading problem for mobile-edge cloud computing in a multi-channel wireless interference environment. We show that it is NP-hard to compute a centralized optimal solution, and hence adopt a game theoretic approach for achieving efficient computation offloading in a distributed manner. We formulate the distributed computation offloading decision making problem among mobile device users as a multi-user computation offloading game. We analyze the structural property of the game and show that the game admits a Nash equilibrium and possesses the finite improvement property. We then design a distributed computation offloading algorithm that can achieve a Nash equilibrium, derive the upper bound of the convergence time, and quantify its efficiency ratio over the centralized optimal solutions in terms of two important performance metrics. We further extend our study to the scenario of multi-user computation offloading in the multi-channel wireless contention environment. Numerical results corroborate that the proposed algorithm can achieve superior computation offloading performance and scale well as the user size increases.

With increasing popularity of location-based services (LBSs), there have been growing concerns for location privacy. To protect location privacy in a LBS, mobile users in physical proximity can work in concert to collectively change their pseudonyms, in order to hide spatial-temporal correlation in their location traces. In this study, we leverage the social tie structure among mobile users to motivate them to participate in pseudonym change. Drawing on a social group utility maximization (SGUM) framework, we cast users' decision making of whether to change pseudonyms as a socially-aware pseudonym change game (PCG). The PCG further assumes a general anonymity model that allows a user to have its specific anonymity set for personalized location privacy. For the SGUM-based PCG, we show that there exists a socially-aware Nash equilibrium (SNE), and quantify the system efficiency of the SNE with respect to the optimal social welfare. Then we develop a greedy algorithm that myopically determines users' strategies, based on the social group utility derived from only the users whose strategies have already been determined. It turns out that this algorithm can efficiently find a Pareto-optimal SNE with social welfare higher than that for the socially-oblivious PCG, pointing out the impact of exploiting social tie structure. We further show that the Pareto-optimal SNE can be achieved in a distributed manner.

The recent development of social services tightens wireless users' social relationships and encourages them to generate more data traffic under network effect. This boosts the demand for wireless services yet may challenge the limited wireless capacity. To fully exploit this opportunity, we study mobile users' data usage behaviors by jointly considering the network effect based on their social relationships in the social domain and the congestion effect in the physical wireless domain. Accordingly, we develop a Stackelberg game for problem formulation: In Stage I, a wireless provider first decides the data pricing to all users to maximize its revenue, and then in Stage II users observe the price and decide data usage subject to mutual interactions under both network and congestion effects. We analyze the two-stage game using backward induction. For Stage II, we first show the existence and uniqueness of a user demand equilibrium (UDE). Then we propose a distributed update algorithm for users to reach the UDE. Furthermore, we investigate the impacts of different parameters on the UDE. For Stage I, we develop an optimal pricing algorithm to maximize the wireless provider's revenue. We evaluate the performance of our proposed algorithms by numerical studies using real data, and thereby draw useful engineering insights for the operation of wireless providers.

AS-level end-to-end paths are of great value for ISPs and a variety of network applications. Although tools like traceroute may reveal AS paths, they require the permission to access source hosts and introduce additional probing traffic, which is not feasible in many applications. In contrast, AS path inference based on BGP control plane data and AS relationship information is a more practical and cost-effective approach. However, this approach suffers from a limited accuracy and high traffic, especially when AS paths are long. In this paper, we bring a new angle to the AS path inference problem by exploiting the metrical tree-likeness or low hyperbolicity of the Internet, part of the complex network properties of the Internet. We show that such property can generate a new constraint that narrows down the searching space of possible AS paths to a much smaller size. Based on this observation, we propose two new AS path inference algorithms, namely HyperPath and Valley-free HyperPath. With intensive evaluations on AS paths from real-world BGP Routing Information Bases, we show that the proposed new algorithms can achieve superior performance, in particular, when AS paths are long paths. We demonstrate that our algorithms can significantly reduce inter-AS traffic for P2P applications with an improved AS path prediction accuracy.

Device‐to‐device (D2D) communication has recently attracted much research attention because of its potential to increase the capacity of cellular networks. Most existing works aim to maximise the overall system throughput (system‐centric), which ignores the actual traffic demands of D2D users. In this study, the authors consider user‐centric relay assisted D2D communications where D2D users have different evaluations for the significance of every unit of increased data rate. By considering the traffic demands of D2D users, the authors propose a Vickrey–Clarke–Groves auction based relay allocation mechanism (ARM) in which every D2D user submits a bid to the basestation (BS). The submitted bids indicate D2D users’ valuation on every unit of the increased data rate. The BS then allocates relays to D2D users by maximising the social welfare of D2D users while maintaining a predefined data rate requirement for cellular users. A payment scheme to charge D2D users for using relays is designed, and the authors show that the auction is truthful. The authors also extend the results to a general case and provide a general ARM accordingly. Extensive simulation results are provided to demonstrate the performance of the proposed mechanisms.

A key feature of wireless communications is the spatial reuse. However, the spatial aspect is not yet well understood for the purpose of designing efficient spectrum sharing mechanisms. In this paper, we propose a framework of spatial spectrum access games on directed interference graphs, which can model quite general interference relationship with spatial reuse in wireless networks. We show that a pure Nash equilibrium exists for the two classes of games: (1) any spatial spectrum access games on directed acyclic graphs, and (2) any games satisfying the congestion property on directed trees and directed forests. Under mild technical conditions, the spatial spectrum access games with random backoff and Aloha channel contention mechanisms on undirected graphs also have a pure Nash equilibrium. We also quantify the price of anarchy of the spatial spectrum access game. We then propose a distributed learning algorithm, which only utilizes users' local observations to adaptively adjust the spectrum access strategies. We show that the distributed learning algorithm can converge to an approximate mixed-strategy Nash equilibrium for any spatial spectrum access games. Numerical results demonstrate that the distributed learning algorithm achieves up to 100 percent performance improvement over a random access algorithm.

Dynamic spectrum sharing is a promising technology for improving the spectrum utilization. In this paper, we study how secondary users can share the spectrum in a distributed fashion based on social imitations. The imitation-based mechanism leverages the social intelligence of the secondary user crowd and only requires a low computational power for each individual user. We introduce the information sharing graph to model the social information sharing relationship among the secondary users. We propose an imitative spectrum access mechanism on a general information sharing graph such that each secondary user first estimates its expected throughput based on local observations, and then imitates the channel selection of another neighboring user who achieves a higher throughput. We show that the imitative spectrum access mechanism converges to an imitation equilibrium, where no beneficial imitation can be further carried out on the time average. Numerical results show that the imitative spectrum access mechanism can achieve efficient spectrum utilization and meanwhile provide good fairness across secondary users.

Thanks to the convergence of pervasive mobile communications and fast-growing online social networking, mobile social networking is penetrating into our everyday life. Aiming to develop a systematic understanding of mobile social networks, in this paper we exploit social ties in human social networks to enhance cooperative device-to-device (D2D) communications. Specifically, as handheld devices are carried by human beings, we leverage two key social phenomena, namely social trust and social reciprocity, to promote efficient cooperation among devices. With this insight, we develop a coalitional game-theoretic framework to devise social-tie-based cooperation strategies for D2D communications. We also develop a network-assisted relay selection mechanism to implement the coalitional game solution, and show that the mechanism is immune to group deviations, individually rational, truthful, and computationally efficient. We evaluate the performance of the mechanism by using real social data traces. Simulation results corroborate that the proposed mechanism can achieve significant performance gain over the case without D2D cooperation.

Mobile cloud computing is envisioned as a promising approach to augment computation capabilities of mobile devices for emerging resource-hungry mobile applications. In this paper, we propose a game theoretic approach for achieving efficient computation offloading for mobile cloud computing. We formulate the decentralized computation offloading decision making problem among mobile device users as a decentralized computation offloading game. We analyze the structural property of the game and show that the game always admits a Nash equilibrium. We then design a decentralized computation offloading mechanism that can achieve a Nash equilibrium of the game and quantify its efficiency ratio over the centralized optimal solution. Numerical results demonstrate that the proposed mechanism can achieve efficient computation offloading performance and scale well as the system size increases.

Cooperative networking is a promising technology to meet the rapidly growing demand of mobile data traffic. To stimulate effective and trustworthy user cooperation, we leverage the knowledge of the social tie structure among mobile users and develop a social trust based cooperative D2D relaying framework, which takes into account both physical distances and social distances among users. Based on (finite-horizon) optimal stopping theory, we derive the optimal social aware relay selection strategy, which strikes a balance between performance gain and relay probing cost. We further show that the optimal stopping policy for social aware relay selection exhibits a stage-dependent threshold structure that has a monotonically non-increasing property. Numerical results demonstrate that the proposed mechanism can yield significant throughput gain over the direct transmission scheme.

In this paper, we propose SoCast — a cooperative video multicast framework to stimulate effective cooperation among mobile users (clients), by leveraging two types of important social ties, i.e., social trust and social reciprocity. By using SoCast, clients can form groups to restore incomplete video frames by obtaining missing packets from other clients, according to the unique video encoding structure. In return, the user perception video quality of mobile video multicast can be improved. Specifically, we first cast the problem of social ties based group formation among clients as a coalitional game, and then devise a distributed algorithm to obtain the core solution (group formation) for the formulated coalitional game. Further, a resource allocation mechanism is proposed for the base station to handle radio resource requests from client groups. Extensive numerical studies with real video traces corroborate the significant performance gain by using the SoCast.

In this paper, we develop a social group utility maximization (SGUM) framework for cooperative networking that takes into account both social relationships and physical coupling among users. Specifically, instead of maximizing its individual utility or the overall network utility, each user aims to maximize its social group utility that hinges heavily on its social ties with other users. We show that this framework provides rich modeling flexibility and spans the continuum space between non-cooperative game and network utility maximization (NUM) - two traditionally disjoint paradigms for network optimization. Based on this framework, we study an important application in database assisted spectrum access. We formulate the distributed spectrum access problem among white-space users with social ties as a SGUM game. We show that the game is a potential game and always admits a social-aware Nash equilibrium. We also design a distributed spectrum access algorithm that can achieve the social-aware Nash equilibrium of the game and quantify its performance gap. We evaluate the performance of the SGUM solution using real social data traces. Numerical results demonstrate that the performance gap between the SGUM solution and the NUM (social welfare optimal) solution is at most 15%.

This paper studies secret key establishment between two adjacent mobile nodes, which is crucial for securing emerging device-to-device (D2D) communication. As a promising method, cooperative key generation allows two mobile nodes to select some common neighbors as relays and directly extract a secret key from the wireless channels among them. A challenging issue that has been overlooked is that mobile nodes are often self-interested and reluctant to act as relays without adequate reward in return. We propose SYNERGY, a game-theoretical approach for stimulating cooperative key generation. The underlying idea of SYNERGY is to partition a group of mobile nodes into disjoint coalitions such that the nodes in each coalition fully collaborate on cooperative key generation. We formulate the group partitioning as a coalitional game and design centralized and also distributed protocols for obtaining the core solution to the game. The performance of SYNERGY is evaluated by extensive simulations.

Smart meters, designed for information collection and system monitoring in smart grid, report fine-grained power consumption to utility providers. With these highly accurate profiles of energy usage, however, it is possible to identify consumers' specific activity or behavior patterns, thereby giving rise to serious privacy concerns. In this paper, this concern is addressed by using battery energy storage. Beyond privacy protection, batteries can also be used to cut down the electricity bill. From a holistic perspective, a dynamic optimization framework is designed for consumers to strike a tradeoff between the smart meter data privacy and the electricity bill. In general, a major challenge in solving dynamic optimization problems lies in the need of the knowledge of the future electricity consumption events. By exploring the underlying structure of the original problem, an equivalent problem is derived, which can be solved by using only the current observations. An online control algorithm is then developed to solve the equivalent problem based on the Lyapunov optimization technique. To overcome the difficulty of solving a mixed-integer nonlinear program involved in the online control algorithm, the problem is further decomposed into multiple cases and the closed-form solution to each case is derived accordingly. It is shown that the proposed online control algorithm can optimally control the battery operations to protect the smart meter data privacy and cut down the electricity bill, without the knowledge of the statistics of the time-varying load requirement and the electricity price processes. The efficacy of the proposed algorithm is demonstrated through extensive numerical evaluations using real data.

Today's wireless networks are increasingly crowded with an explosion of wireless users, who have greater and more diverse quality of service (QoS) demands than ever before. However, the amount of spectrum that can be used to satisfy these demands remains finite. This leads to a great challenge for wireless users to effectively share the spectrum to achieve their QoS requirements. This paper presents a game theoretic model for spectrum sharing, where users seek to satisfy their QoS demands in a distributed fashion. Our spectrum sharing model is quite general, because we allow different wireless channels to provide different QoS, depending upon their channel conditions and how many users are trying to access them. Also, users can be highly heterogeneous, with different QoS demands, depending upon their activities, hardware capabilities, and technology choices. Under such a general setting, we show that it is NP hard to find a spectrum allocation which satisfies the maximum number of users' QoS requirements in a centralized fashion. We also show that allowing users to self-organize through distributed channel selections is a viable alternative to the centralized optimization, because better response updating is guaranteed to reach a pure Nash equilibria in polynomial time. By bounding the price of anarchy, we demonstrate that the worst case pure Nash equilibrium can be close to optimal, when users and channels are not very heterogenous. We also extend our model by considering the frequency spatial reuse, and consider the user interactions as a game upon a graph where players only contend with their neighbors. We prove that better response updating is still guaranteed to reach a pure Nash equilibrium in this more general spatial QoS satisfaction game.

In this paper, we study the distributed spectrum sharing problem with spatial reuse and without explicit message passing. We propose tow novel threshold-based decentralized spatial spectrum access algorithms, which do not require information exchange among secondary users and channel switching at the equilibrium state. Moreover, we show that the proposed algorithms can converge to either an approximate Nash equilibrium or an approximate Pareto optimum, based on the different threshold designs. Numerical results show that the performance loss of the proposed algorithms is less than 10%, compared with the centralized optimal solution.

Today's wireless networks are facing tremendous growth and many applications have more demanding quality of service (QoS) requirements than ever before. However, there is only a finite amount of wireless resources (such as spectrum) that can be used to satisfy these demanding requirements. We present a general QoS satisfaction game framework for modeling the issue of distributed spectrum sharing to meet QoS requirements. Our study is motivated by the observation that finding globally optimal spectrum sharing solutions with QoS guarantees is NP hard. We show that the QoS satisfaction game has the finite improvement property, and the users can self-organize into a pure Nash equilibrium in polynomial time. By bounding the price of anarchy, we demonstrate that the worst case pure Nash equilibrium can be close to the global optimal solution when users' QoS demands are not too diverse.

Thanks to the convergence of pervasive mobile communications and fast-growing online social networking, mobile social networking is penetrating into our everyday life. Aiming to develop a systematic understanding of the interplay between social structure and mobile communications, in this paper we exploit social ties in human social networks to enhance cooperative device-to-device communications. Specifically, as hand-held devices are carried by human beings, we leverage two key social phenomena, namely social trust and social reciprocity, to promote efficient cooperation among devices. With this insight, we develop a coalitional game theoretic framework to devise social-tie based cooperation strategies for device-to-device communications. We also develop a network assisted relay selection mechanism to implement the coalitional game solution, and show that the mechanism is immune to group deviations, individually rational, and truthful. We evaluate the performance of the mechanism by using real social data traces. Numerical results show that the proposed mechanism can achieve up-to 122% performance gain over the case without D2D cooperation.

Efficient distributed spectrum sharing mechanism is crucial for improving the spectrum utilization. The spatial aspect of spectrum sharing, however, is less understood than many other aspects. In this paper, we generalize a recently proposed spatial congestion game framework to design efficient distributed spectrum access mechanisms with spatial reuse. We first propose a spatial channel selection game to model the distributed channel selection problem with fixed user locations. We show that the game is a potential game, and develop a distributed learning mechanism that converges to a Nash equilibrium only based on users' local observations. We then formulate the joint channel and location selection problem as a spatial channel selection and mobility game, and show that it is also a potential game. We next propose a distributed strategic mobility algorithm, jointly with the distributed learning mechanism, that can converge to a Nash equilibrium.

According to FCC's ruling for white-space spectrum access, white-space devices are required to query a database to determine the spectrum availability. In this paper, we study the database-assisted distributed white-space access point (AP) network design. We first model the cooperative and non-cooperative channel selection problems among the APs as the system-wide throughput optimization and non-cooperative AP channel selection games, respectively, and design distributed AP channel selection algorithms that achieve system optimal point and Nash equilibrium, respectively. We then propose a state-based game formulation for the distributed AP association problem of the secondary users by taking the cost of mobility into account. We show that the state-based distributed AP association game has the finite improvement property, and design a distributed AP association algorithm that can converge to a state-based Nash equilibrium. Numerical results show that the algorithm is robust to the perturbation by secondary users' dynamical leaving and entering the system.

In this paper, we design distributed spectrum access mechanisms with both complete and incomplete network information. We propose an evolutionary spectrum access mechanism with complete network information, and show that the mechanism achieves an equilibrium that is globally evolutionarily stable. With incomplete network information, we propose a distributed learning mechanism, where each user utilizes local observations to estimate the expected throughput and learns to adjust its spectrum access strategy adaptively over time. We show that the learning mechanism converges to the same evolutionary equilibrium on the time average. Numerical results show that the proposed mechanisms achieve up to 35 percent performance improvement over the distributed reinforcement learning mechanism in the literature, and are robust to the perturbations of users' channel selections.

We propose a dynamic spectrum access scheme where secondary users cooperatively recommend "goodâ channels to each other and access accordingly. We formulate the problem as an average reward-based Markov decision process. We show the existence of the optimal stationary spectrum access policy and explore its structure properties in two asymptotic cases. Since the action space of the Markov decision process is continuous, it is difficult to find the optimal policy by simply discretizing the action space and use the policy iteration, value iteration, or Q-learning methods. Instead, we propose a new algorithm based on the model reference adaptive search method and prove its convergence to the optimal policy. Numerical results show that the proposed algorithms achieve up to 18 and 100 percent performance improvement than the static channel recommendation scheme in homogeneous and heterogeneous channel environments, respectively, and is more robust to channel dynamics.

In this paper, we present a novel framework for spectrum sharing in cognitive radio networks. The secondary users (SUs) can share the spectrum resource with primary users (PUs) in a cooperative manner, where PUs trade their information and surplus resource, and SUs access the primary spectrum intelligently based on SUs' heterogeneous demands and PUs' resource prices. After paying PUs a subscription fee for the spectrum information, SUs become spectrum-aware and avoid the overhead on spectrum sensing. During SUs' channel access, PUs further charge SUs based on the amount of resource taken by SUs. We model this sharing problem in a flexible time-slotted structure, where SUs' decisions include the selection of proper transmission channel and slot length to meet their demands. This joint decision problem is studied as a spectral temporal allocation game. We prove the existence of a Nash equilibrium and design a strategy update process which can converge to an equilibrium.

According to FCC's ruling for white-space spectrum access, white-space devices are required to query a database to determine the spectrum availability. In this paper, we adopt a game theoretic approach for the database-assisted white-space access point (AP) network design. We first model the channel selection problem among the APs as a distributed AP channel selection game, and design a distributed AP channel selection algorithm that achieves a Nash equilibrium. We then propose a state-based game formulation for the distributed AP association problem of the secondary users by taking the cost of mobility into account. We show that the state-based distributed AP association game has the finite improvement property, and design a distributed AP association algorithm can converge to a state-based Nash equilibrium. Numerical results show that the algorithm is robust to the perturbation by secondary users' dynamical leaving and entering the system.

A key feature of wireless communications is the spatial reuse. However, the spatial aspect is not yet well understood for the purpose of designing efficient spectrum sharing mechanisms. In this paper, we propose a framework of spatial spectrum access games on directed interference graphs, which can model quite general interference relationship with spatial reuse in wireless networks. We show that a pure strategy equilibrium exists for the two classes of games: (1) any spatial spectrum access games on directed acyclic graphs, and (2) any games satisfying the congestion property on directed trees and directed forests. Under mild technical conditions, the spatial spectrum access games with random backoff and Aloha channel contention mechanisms on undirected graphs also have a pure Nash equilibrium. We then propose a distributed learning algorithm, which only utilizes users' local observations to adaptively adjust the spectrum access strategies. We show that the distributed learning algorithm can converge to an approximate mixed-strategy Nash equilibrium for any spatial spectrum access games. Numerical results demonstrate that the distributed learning algorithm achieves up to 100% performance improvement over a random access algorithm.

In this paper, we consider the open spectrum access mechanism design with both complete and incomplete network information in a many-users regime. We propose an evolutionary spectrum access mechanism with complete network information, and show that the mechanism achieves an equilibrium that is both evolutionarily stable and globally stable. With incomplete network information, we propose a distributed learning mechanism, where each user utilizes local observations to estimate the channel quality and learns to adjust its spectrum access strategy adaptively over time. Numerical results show that the proposed mechanisms achieve efficient spectrum sharing among the users, and are robust to the perturbations of users' channel selections.

Proper patrol route planning increases the effectiveness of police patrolling and improves public security. In this paper we present a new approach for the real-time patrol route planning in a dynamic environment. We first build a mathematic framework, and then propose a fast algorithm developed from the Cross Entropy method to meet the real-time computation requirement needed for many applications. In addition, as the randomness is an important factor for practices, the entropy concept is used for designing the randomized patrol routes schedule strategy. Numerical studies demonstrate that the approach has fast convergence property and is efficient in dynamic patrol environment.

Public security is a key concern around the world. Efficient patrol strategy increases the effectiveness of police patrolling and improves public security. In this paper we propose a new general security measure by defining the security level function. Based on this, we present the balanced patrol districting solution for the multiple units assignment problem. For the patrol routing problem in a patrol district, we first formulate the patrol routing process as a graph-based Markov decision process, and then propose an ε-optimal patrol routing strategy to deal with the curse of dimensionality. The strategy is derived based on the concept of ε-optimal horizon approximation. Numerical studies demonstrate that the strategy is adaptive to the generalized security measure by security level function, and has significant performance improvement over the referenced strategies in previous works. In addition, as the randomness is an important factor for practices, we design the randomized patrol routing strategy on the basis of the randomized exploration method in the Reinforcement Learning.