Large Language Models (LLMs) exhibit pronounced memory-bound characteristics during inference due to High Bandwidth Memory (HBM) bandwidth constraints. In this paper, we propose an L2 Cache-oriented asynchronous KV Cache prefetching method to break through the memory bandwidth bottleneck in LLM inference through computation-load overlap. By strategically scheduling idle memory bandwidth during active computation windows, our method proactively prefetches required KV Cache into GPU L2 cache, enabling high-speed L2 cache hits for subsequent accesses and effectively hiding HBM access latency within computational cycles. Extensive experiments on NVIDIA H20 GPUs demonstrate that the proposed method achieves 2.15× improvement in attention kernel efficiency and up to 1.97× end-to-end throughput enhancement, surpassing state-of-the-art baseline FlashAttention-3. Notably, our solution maintains orthogonality to existing optimization techniques and can be integrated with current inference frameworks, providing a scalable latency-hiding solution for next-generation LLM inference engines.

Synthesizing high-quality tabular data is essential for privacy-preserving data analysis. However, this task remains challenging due to two key factors: (1) distribution complexity : imbalanced and skewed data make it challenging to learn the data distribution accurately; and (2) semantic coherence : implicit relationships and logical dependencies among fields must be preserved to ensure valid and meaningful synthetic samples. To address these issues, we propose SynDiSC, a high-quality tabular data synthesis approach that enforces both distributional and semantic consistency. It comprises three core designs: (1) a distribution-aware encoding that effectively handles heterogeneous data types and complex distributions; (2) a multi-dimensional semantic conditioning that leverages multi-dimensional conditional dependencies to enforce semantic validity during generation; and (3) a conditional consistency controller that guides the generator to produce diverse samples satisfying multiple conditional constraints while mitigating mode collapse. Extensive experiments on datasets from various application domains demonstrate that SynDiSC significantly improves data quality, conditional controllability, and downstream task performance compared to state-of-the-art methods. Our code and data samples are open-sourced in the GitHub repository. https://github.com/Knightz9/SynDiSC.

Mobile network operators (e.g., China Mobile, Verizon) are significant for providing communication services and collecting massive amounts of data. However, operators are increasingly concerned about customer data breaches involving third-party application providers (e.g., Tencent, Apple, Netflix). This concern is particularly aggravated when anonymous datasets shared with third-party providers or publicly released can be linked to user data compromised in breaches, leading to severe re-identification attacks and privacy threats. However, comprehensive methods for identifying such privacy risks on a large scale are lacking due to limited networking behavioral data. To address this, we aim to measure the re-identification privacy risk associated with sharing or releasing cellular traces amidst data breaches. Based on the analysis of key privacyimpacting features in traffic usage and base station association data, we propose a novel re-identification method, SURE, which learns similarities between cellular traces to classify if traces belong to the same user. Extensive experiments on a largescale dataset of 10,000 users over four months demonstrate SURE's superior performance, with AUC scores exceeding 0.9. Our findings reveal significant re-identification risks in data sharing/release, influenced by data scale and user attributes, corroborated by a public dataset.

Understanding the nature of user-device interactions (e.g., who is using the device and what he/she is doing with it) is critical for many applications including time management, user profiles, and privacy protection. However, in scenarios where mobile devices are shared among family members or multiple employees in a company, conventional account-based statistics are not meaningful. This poses an even bigger problem when dealing with sensitive data. Moreover, fingerprint readers and front-facing cameras were not designed to continuously identify users. In this study, we developedMagPrint++, a novel approach to fingerprint users based on unique patterns in the electromagnetic (EM) signals associated with the specific use patterns of users. Initial experiments showed that time-varying EM patterns are unique to individual users. They are also temporally and spatially consistent, which makes them suitable for fingerprinting.MagPrint++has a number of advantages over existing schemes: i) Non-intrusive fingerprinting, ii) implementation both on COTS mobile phones and a small and easy-to-deploy device, and iii) high accuracy thanks to the proposed classification algorithm. In experiments involving 30 users,MagPrint++achieves$94.3\%$accuracy in classifying users from these traces, which represents a$10.9\%$improvement over the state-of-the-art classification method.

On-device Large Language Model (LLM) inference enables private, personalized AI but faces memory constraints. Despite memory optimization efforts, scaling laws continue to increase model sizes and memory pressure. In this paper, we revisit the core memory bottlenecks in on-device LLM inference and conduct a comprehensive analysis of mainstream optimization techniques. We uncover several overlooked inefficiencies: (1) model weights, not KV caches, dominate memory usage; (2) weight sparsity remains underutilized; (3) OS-level memory behaviors cause redundancy; and (4) naive weight loading leads to excessive memory residency. To address these challenges, we propose H2O, a heterogeneity-aware hierarchical orchestration framework for memory-efficient on-device LLM inference. H2O introduces three key techniques, including hierarchical weight orchestration to reduce redundant memory retention, zero copy I/O–compute parallelism for safe and efficient memory reuse, and heterogeneity-aware inference planning to adapt to diverse mobile hardware constraints. Extensive experimental results show that H2O reduces peak memory usage by up to 60%, eliminates out-of-memory (OOM) failures for 7B–13B models, and improves inference latency by 34%–94% under tight memory budgets. We open-source our implementation at: https://github.com/ccfeiker/H2O.

The rise of vehicular edge computing (VEC) enables vehicles to offload resource-intensive tasks to roadside units (RSUs), improving efficiency and reducing latency. However, high mobility, dynamic resource availability, and heterogeneous quality-of-experience (QoE) requirements make task offloading and coordination highly challenging. In this paper, we investigate the heterogeneous task offloading problem in dynamic VEC environments by proposing a two-stage optimization framework, TOVEC. Our TOVEC decouples the spatio-temporally coupled decision space into two tractable subproblems and solves it with a two-stage design. In the first stage, we employ a TD3-based deep reinforcement learning algorithm to handle RSU-channel access decisions under dynamic network conditions. In the second stage, we formulate the interaction between vehicular users (VUs) and RSUs as a Stackelberg game, enabling joint task scheduling and dynamic pricing that balances VUs’ QoE optimization and RSUs’ revenue maximization. We theoretically prove the existence of a Stackelberg equilibrium and validate our TOVEC using real-world vehicular traces. The experimental results show that our method has improved the QoE (measured by task delay and energy cost) of VU and the benefits of RSU by 6.8%-41.3% and 2.2%-117.4%, respectively, compared with the baseline schemes.

Powerline inspection with UAVs significantly improves efficiency and safety in remote areas. However, the lack of cellular infrastructure necessitates the use of UAV-mesh networks, whose deployment presents challenges in jointly optimizing coverage, node load, robustness, and cost under complex terrain constraints. In this paper, we investigate the computational complexity of this deployment problem by formulating it as a multi-objective optimization task and proving its NP-hardness. To address this, we presentU-Mesh+, aterrain-aware, robust, and cost-efficientdeployment framework that integrates four key components: (i) identifying line-of-sight and non-line-of-sight links to model terrain-induced communication constraints; (ii)NetConsfor cost-effective coverage and connectivity network topology construction; (iii)NetOptfor network resilience and balance node-level load improvement without extra cost; and (iv)NetEnhfor service availability enhancement via targeted local refinements. We implementU-Mesh+in a real-world 270km2mountainous forest with 174 power towers and 48km of transmission lines. Extensive experiments demonstrate its efficacy in terms of deployment cost and network performance. On-site network data from the deployed wireless network further validate its effectiveness and scalability under real-world conditions.

UAV-assisted task execution is a promising approach to powerline inspections in remote areas where no cellular network infrastructure exists for inspection data transmission. In this paper, we investigate UAV-mesh network deployment in remote areas to empower UAV-assisted powerline inspection, which is challenging considering a mountainous environment with no power supply. Particularly, given the locations of a set of power towers, we first formulate the UAV-mesh network deployment problem with connectivity and coverage constraints, which is NP-hard. Then, we propose U-Mesh, which is a cost-effective and load-balanced deployment scheme. To be specific, U-Mesh integrates three components, i.e., link identification: identifying the link conditions between power towers based on geographical barriers, NetCons: conducting local search to gradually obtain a cost-effective initial mesh nodes with connectivity and coverage constraints, and NetOpti: optimizing the initial mesh node positions to improve the mesh load and robustness from the perspectives of overall network structure. Finally, we implement U-Mesh in a 270 km2mountain forest area, and demonstrate its efficacy in terms of both deployment cost and network performance via extensive evaluations.

The deployment of high-accuracy low-distinction and multi-scale object detection models on resource-constrained edge devices is essential for ubiquitous intelligent applications, from autonomous obstacle avoidance to anomaly object recognition. However, these models' computational burden, energy consumption, and memory footprint pose significant challenges for distributed and pervasive systems. In this paper, we propose HiPOD, a hierarchical pruning framework designed to prune low-distinction and multi-scale object detection models by jointly learning layer-wise and path-wise pruning strategies. HiPOD balances the accuracy-efficiency trade-off through two novel components: Layer-Adaptive Ratio Learning, named AdaLR, which leverages network structural characteristics and a reinforcement learning-based action-feedback mechanism to adaptively generate balanced layer-wise pruning ratios; and Genetic Path Optimization, named GenPath, which employs crossover and mutation operations to optimize inter-layer kernel pruning paths, preserving critical semantic and spatial information. Extensive experiments on three benchmark datasets demonstrate the effectiveness of HiPOD, with only a 3 % drop in mAP50 and a 2 % drop in mAP50:95 compared to the full models. The ablation studies and impact analysis further validate the effectiveness of each module and highlight the robustness of our framework. Furthermore, evaluations on an edge device demonstrate the practicality of the proposed solution for powerline inspection.

UAV-assisted inspection is critical for modern power grid maintenance, enhancing efficiency and safety in remote areas. However, automatically designing UAV inspection trajectories is challenging due to the cluttered inspection environments, small inspection targets, and pervasive obstacles. We propose Auto-UIT, a novel method for generating inspection trajectories in noisy, sparse, and complex 3D point cloud. Auto-UIT has three core techniques: (1) A local structure-enhanced pylon segmentation, which accurately segments pylons, power lines, and surroundings in noisy point cloud for effective inspection target identification and trajectory planning. (2) A 3D fingerprint-based pylon type recognition that compensates for point cloud sparsity to complete missing inspection targets based on the pylon type. (3) An adaptive trajectory generation that samples positions in response to diverse pylon orientations and pervasive environmental obstacles, ensuring UAV operational safety. Our experiments on a real-world dataset across four distinct areas demonstrate that Auto-UIT outperforms existing baseline methods in all three tasks. Furthermore, a four-month deployment in a power grid inspection system—covering a 270 km2 primary mountainous area—yielded an expert first-review acceptance rate of 91.86% for the generated trajectories, and reduced design time by an average of 88.19% compared to manual methods, significantly improving inspection efficiency.

UAV-assisted inspection is critical for modern power grid maintenance, enhancing efficiency and safety in remote areas. However, automatically designing UAV inspection trajectories is challenging due to the cluttered inspection environments, small inspection targets, and pervasive obstacles. We propose a novel method for generating inspection trajectories in noisy, sparse, and complex 3D point cloud. It has three core techniques: (1) A local structure-enhanced pylon segmentation, which accurately segments pylons, power lines, and surroundings in noisy point cloud for effective inspection target identification and trajectory planning. (2) A 3D fingerprint-based pylon type recognition that compensates for point cloud sparsity to complete missing inspection targets based on the pylon type. (3) An adaptive trajectory generation that samples positions in response to diverse pylon orientations and pervasive environmental obstacles, ensuring UAV operational safety. A four-month deployment in a power grid inspection system—covering a 270 km2 primary mountainous area—yielded an expert first-review acceptance rate of 91.86% for the generated trajectories, and reduced design time by an average of 88.19% compared to manual methods, significantly improving inspection efficiency. Demo video and dataset are available at https://ljhe006.github.io/autouit/.

Unmanned sanitation vehicles (USVs) promise cleaner cities, yet efficiently coordinating multiple USVs in large urban areas remains challenging due to constraints such as limited waste capacity and battery life. In this demo, we present MRTC, a multi-robot task cooperation system. First, Dynamic Task Assignment employs an Actor-Critic policy within a Markov decision framework to allocate cleaning tasks and decide the required number of USVs. Second, Single-USV Path Planning refines each route via a fast two-layer iterative search. Over an eight-month real-world deployment in three urban testbeds, our MRTC system markedly improved cleaning efficiency while lowering operating costs. Operating over a combined 10,775 km of routes per month, the system achieved average monthly savings of 20,575 kWh of energy and 2,744 labour hours. A demonstration video is available at https://llq978.github.io/Demo/.

The number of Internet of Things (IoT) devices has increased rapidly in recent years, but lack effective methods to integrate their computational power. In this article, we propose NC-Load, which couples IoT devices into a multiprocessor system, allowing process scheduling across different devices to share their computing power and improve overall throughput. Specifically, NC-Load consists of three key designs, i.e., remote page fault (RPF), lightweight program cropping, and identical memory layout migration, contributing to three merits compared to existing systems: 1) high storage efficiency: the target device launches the program with a locally stored lightweight icon and leverages RPFs to retrieve the required code/data from the source device; 2) on-demand memory loading: only the required memory portions are transmitted when scheduling programs across different devices, which ensures quick recovery of the program; and 3) consistent memory layout: to ensure consistency of addresses after program offloading, the virtual memory area layout of the source device is migrated to the target device. We implement NC-Load on Linux 6.1 and conduct performance evaluation using unmodified programs and the N-Queens cases. The results demonstrate that NC-Load can achieve superior performance in terms of storage efficiency, program performance, memory usage, and throughput.

The all-reduce operation is critically important for communication-intensive workloads emerging at the convergence of High-Performance Computing (HPC) and Internet of Things (IoT) applications. However, existing optimization efforts primarily concentrate on offloading the all-reduce onto network switches, known as In-Network Aggregation, which are incompatible with switchless torus topologies. Driven by our systematic analysis, we identified two key factors that impact the performance of the standard ring all-reduce operation: i. The all-reduce computation process frequently interrupts the GPU/CPU’s computation tasks; ii. The GPU/CPU is, in fact, indifferent to intermediate computational results. Based on this insight, we propose DirectReduce, a fully offloading ring all-reduce architecture that is comprised of three components: (i) the GateKeeper module, responsible for evaluating outgoing data to decide its progression-either directing it to the Protocol Engine for packetization or intercepting it for reduction (e.g., sum, maximum); (ii) the DataDirector module, which classifies the incoming data either is for intermediate result reduction or final result storage; and (iii) the ComputeEnhancer module, designed to execute reduction operations directly on the SmartNIC. Extensive simulation results show that DirectReduce can reduce the ring all-reduce latency by up to 1.98X in a ring (1D-torus) topology, 1.97X in a 3D-torus topology, and 1.75X in a 6D-torus topology compared to the standard ring all-reduce.

Syndrome-based early epidemic warning plays a vital role in preventing and controlling unknown epidemic outbreaks. It monitors the frequency of each syndrome, issues a warning if some frequency is aberrant, identifies potential epidemic outbreaks, and alerts governments as early as possible. Existing systems adopt a cloud-assisted paradigm to achieve cross-facility statistics on the syndrome frequencies. However, in these systems, all symptom data would be directly leaked to the cloud, which causes critical security and privacy issues. In this paper, we first analyze syndrome-based early epidemic warning systems and formalize two security notions, i.e., symptom confidentiality and frequency confidentiality, according to the inherent security requirements. We propose extsf{EpiOracle}, a cross-facility early warning scheme for unknown epidemics. EpiOracle ensures that the contents and frequencies of syndromes will not be leaked to any unrelated parties; moreover, our construction uses only a symmetric-key encryption algorithm and cryptographic hash functions (e.g., [CBC]AES and SHA-3), making it highly efficient. We formally prove the security of EpiOracle in the random oracle model. We also implement an EpiOracle prototype and evaluate its performance using a set of real-world symptom lists. The evaluation results demonstrate its practical efficiency.

In this paper, we investigate the problem of enhancing cooperative LiDAR-based perception accuracy in vehicular edge networks. The key to solving this problem is the selection of connected and autonomous vehicles (CAVs) that can collectively provide maximum perception performance. In specific, extensive motivating experiments on an open benchmark dataset are conducted, which reveal that the cooperative perception accuracy is a submodular combination of selected CAVs, and such selection is non-trivial due to high vehicular mobility as well as unstable vehicular network conditions. Then, we develop an Edge coordinated COoperative Perception (ECOP) framework, taking into account both cooperative vehicle selection and adaptive bandwidth allocation. The novelty of the ECOP design is threefold. First, a new metric named perceptual gain is designed, which properly quantifies the individual perception contributions of each CAV without incurring additional computational overhead. Secondly, an online vehicle selection strategy, which utilizes continual learning to assess the perceptual gain of each CAV, is devised. Theoretical analysis indicates that the proposed vehicle selection strategy can achieve asymptotically diminishing learning regret, highlighting its effectiveness in adapting to vehicular mobility. Finally, an optimal bandwidth allocation method is proposed, which can adapt to heterogeneous and unstable vehicular network conditions. Simulation results demonstrate that, compared with other benchmarks, ECOP can select vehicle sets with the highest cooperative perception accuracy and ensure real-time perception in the presence of fluctuating bandwidth. Furthermore, a case study is presented to visualize the effectiveness of the proposed ECOP framework.

Mobile contact exhibits user co-traveling events within the same transportation tool, which is crucial for resident profiling, face-to-face interaction detection, etc. In this paper, we investigate urban user mobile contact detection with cellular signaling traces, which is cost-efficient to enable large-scale detection. Specifically, we develop a data collection platform to collect substantial user signaling traces, covering different types of road scenarios within a city. With the collected traces, we perform systematic data analysis to reveal several technical challenges, which are sparsity of signaling trajectory, remote base station noise, and fuzzy matching difficulties. To address challenges, we propose a mobile contact detection method namedMoCo. InMoCoframework, we first conduct data denoising to remove the noise from remote base stations. Then, we devise a spatio-temporal filter to eliminate unlikely mobile contact traces in both spatial and temporal domains, reducing the computational overhead. Finally, we design a detection network that integrates the submodules of data alignment, feature encoder, spatio-temporal representation learner, and user mobile contact detector. Extensive evaluation results demonstrate the superiority ofMoCoin comparison with state-of-the-art baselines. Robust experiments show thatMoCocan work efficiently in different transportation modes and urban densities.

Artificial Intelligence-Generated Content (AIGC) technology is transforming content creation by enabling diverse customized and quality services. However, the limited computing resources on mobile devices hinder the provisioning of AIGC services at scale, pose challenges in guaranteeing user-satisfied content quality requirement. To address these challenges, we first investigate the characteristics of prompt category and inference models in Text-to-Image (T2I) diffusion process. It is observed that, model size, denoising steps, and computing resource, are three deciding factors to image generation utility. Based on this insight, we first design an edge-assisted AIGC service system to efficiently process multi-user T2I generative requests, employing a multi-flow queuing model to capture multi-user dynamics and characterize the impact of diffusion scheduling on service latency. The system schedules the diffusion process of T2I generation across edge-deployed models, balancing service quality and computing resource. To maximize generation utility under resource constraints, we propose a Monte Carlo Tree Search-based diffusion scheduling algorithm embedded with adaptive computing resource allocation subroutine. This algorithm ensures that, resource allocation dynamically adapts to scheduling decisions in real time, enabling an effective trade-off between service quality and latency. Extensive experimental comparison against baseline approaches demonstrates that, the proposed system can enhance the generation utility by up to 7.3$\%$, achieving a 2.9$\%$improvement in quality score and a 33.3$\%$reduction in service latency.

Cellular operators face significant challenges in cutting operating expenses while maintaining the quality of service (QoS) for users due to growing network traffic and dynamic user connections. These challenges are addressed by the cloud radio access network (C-RAN) architecture, which includes a centralized pool of baseband units (BBUs) and distributes them from remote radio heads (RRHs). The key to improving C-RAN performance is to dynamically allocate large-scale RRHs to different BBUs in real time. In this paper, we propose a user behavior-aware RRH-BBU mapping framework to improve the performance of large-scale C-RANs by predicting RRH traffic and users in advance. First, we propose a Multivariate RRH time series Prediction Model (MRPM) that captures the spatio-temporal patterns in the data to predict the traffic volume and the number of users of RRHs, which represents key indicators of RRH connection states. Second, we formulate the RRH-BBU mapping as a Markov decision process problem to optimize cost and QoS by considering BBU utilization, BBU energy consumption, RRH migration frequency, and BBU load balancing. Third, we propose a prediction-based RRH-BBU mapping scheme (PB-RBM) to find the optimal RRH-BBU mapping strategy by leveraging the prediction information of MRPM. In the PB-RBM algorithm, we employ an A3C algorithm to learn the mapping policy and group the RRHs based on a defined popularity metric to reduce the state and action space of the reinforcement learning algorithm. Finally, extensive experiments are conducted on a real-world dataset, and our algorithm is compared with several matching algorithms, such as ACKTR, heuristic, etc., to demonstrate its superiority, especially reducing 17.5% in RMSE compared to the best-performing baseline.

Cellular traffic data hold significant potential for applications such as network planning, traffic prediction, mobility modeling, and personalized recommendations. However, limited data accessibility hinders more open data-driven research. Previous studies have explored data synthesis, while exhibiting flexible limitations in supporting conditional traffic synthesis, and are vulnerable to multidimensional data modeling. In this paper, we present LLMCell, a flexible and effective framework that leverages the arbitrary conditioning and contextual understanding capabilities of the large language model (LLM) to generate high-quality synthetic cellular traffic data. The LLMCell comprises three key components: i) a textual encoder for converting raw cellular traffic data into textual representations, ii) a generative model learner to fine-tune pre-trained LLM based on encoded textual representation for cellular traffic generation, and iii) a synthetic data sampling module for final synthetic data sampling and textual-to-data transformation. Experiments conducted on a large-scale dataset demonstrate the superior fidelity and utility of LLMCell over state-of-the-art baselines, and the synthetic data can effectively preserve user privacy. We release our synthetic dataset to the public to benefit future research in the wireless network community1.

Artificial Intelligence Generated Content (AIGC) services can efficiently satisfy user-specified content creation demands, but the high computational requirements pose various challenges to supporting mobile users at scale. In this paper, we present our design of an edge-enabled AIGC service provisioning system to properly assign computing tasks of generative models to edge servers, thereby improving overall user experience and reducing content generation latency. Specifically, once the edge server receives user requested task prompts, it dynamically assigns appropriate models and allocates computing resources based on features of each category of prompts. The generated contents are then delivered to users. The key to this system is a proposed probabilistic model assignment approach, which estimates the quality score of generated contents for each prompt based on category labels. Next, we introduce a heuristic algorithm that enables adaptive configuration of both generation steps and resource allocation, according to the various task requests received by each generative model on the edge. Simulation results demonstrate that the designed system can effectively enhance the quality of generated content by up to 4.7% while reducing response delay by up to 39.1% compared to benchmarks.

The surging development of Artificial Intelligence-Generated Content (AIGC) marks a transformative era of the content creation and production. Edge servers promise attractive benefits, e.g., reduced service delay and backhaul traffic load, for hosting AIGC services compared to cloud-based solutions. However, the scarcity of available resources on the edge pose significant challenges in deploying generative AI models. In this paper, by characterizing the resource and delay demands of typical generative AI models, we find that the consumption of storage and GPU memory, as well as the model switching delay represented by I/O delay during the preloading phase, are significant and vary across models. These multidimensional coupling factors render it difficult to make efficient edge model deployment decisions. Hence, we present a collaborative edge-cloud framework aiming to properly manage generative AI model deployment on the edge. Specifically, we formulate edge model deployment problem considering heterogeneous features of models as an optimization problem, and propose a model-level decision selection algorithm to solve it. It enables pooled resource sharing and optimizes the trade-off between resource consumption and delay in edge generative AI model deployment. Simulation results validate the efficacy of the proposed algorithm compared with baselines, demonstrating its potential to reduce overall costs by providing feature-aware model deployment decisions.

Electricity theft detection (ETD) aims to promptly identify electricity theft by vigilantly monitoring and analyzing atypical electricity consumption time series. Existing machine learning approaches to ETD demand large training sets, leading to degraded performance when limited training samples are available. In this paper, we introduce FL2ETD, a novel few-shot learning framework to ETD. The framework consists of three core components, i.e., a feature extraction module, a representation module, and a classification module. The feature extraction module processes the electricity consumption behavior of users in both the time and the frequency domains to extract distinctive features and increase the number and the diversity of features. The representation module utilizes contrast learning to pre-train unlabeled electricity consumption data for enhancing feature representation quality. The classification module integrates feature representations for making the final decision in ETD. Extensive experiments demonstrate that FL2ETD exhibits superior performance compared to baselines, and its advantage is significant when the number of available training samples is very small (with only 338 samples).

In this work, we study the problem of dispatching an unmanned aerial vehicle (UAV) for data collection of Internet of Things (IoT) devices, where a UAV departs from a data center, then visits some IoT devices for data collection and finally returns to the data center. Different from most existing works on UAV-enabled data collection, we assume that the UAV’s collection point, i.e., the location where the UAV stays during the data collection process, can be deployed anywhere within the communication range of each IoT device, rather than being assumed to be in a fixed position. This new assumption is motivated by the fact that the collection point has a great impact on both time and energy consumption of the UAV during its data collection tour. Thus, in this work, we focus on minimizing the UAV’s task completion time and energy consumption during a data collection tour, by jointly optimizing the UAV’s collection point for each IoT device, flight trajectory and flight speed. We formulate this problem as a multiobjective optimization problem, which is solved by executing the following three successive steps: 1) we first employ the ant colony optimization (ACO) algorithm to decide the UAV’s visiting order of all IoT devices; 2) we then reduce the searching space of the collection point for each visited IoT device by using geometric theory and reformulate the original problem; and 3) we finally develop an enhanced multiobjective particle swarm optimization (EMOPSO) algorithm by incorporating a novel gbest selection strategy to identify the optimal collection point for each visited IoT device, based on which the corresponding flight trajectory as well as the flight speed is calculated. We refer to the above three-step hybrid algorithm as ACO-EMOPSO-G. Extensive evaluations validate the superiority of ACO-EMOPSO-G in terms of the tradeoff between the UAV’s task completion time and energy consumption, compared with some other data collection approaches.

Nowadays, sign language is becoming increasingly important in people’s daily life. Existing solutions are often based on wireless signals (e.g., acoustic, visible, and WiFi) or wearable sensors to recognize gestures, but they suffer from vulnerability to environmental influences, poor security, and high energy consumption, which prevent them from accurately capturing finger micromovements. In this article, we propose RF-Sign, which uses passive radio-frequency identification (RFID) tags to capture multiple finger micromovements simultaneously to enable sign language support. In particular, two main issues are studied. One is the problem of positional differences when users make the same gesture, and the other is the problem of segmenting consecutive gestures using only empirical thresholding methods and ignoring the existence of differences in thresholds for different gestures. For position differences, we propose position models to normalize the hand’s horizontal rotation angle and radial distance. For segmenting consecutive gestures, we use the received signal strength (RSS) trend of the reference tag to represent the finger micromovements state. The experimental results show that the average accuracy reaches 92.81% under different angles, distances, and other conditions.

Mobile edge computing is a promising computing paradigm enabling mobile devices to offload computation-intensive tasks to nearby edge servers. However, within small-cell networks, the user mobilities can result in uneven spatio-temporal loads, which have not been well studied by considering adaptive load balancing, thus limiting the system performance. Motivated by the data analytics and observations on a real-world user association dataset in a large-scale WiFi system, in this paper, we investigate the mobility-aware online task offloading problem with adaptive load balancing to minimize the total computation costs. However, the problem is intractable directly without prior knowledge of future user mobility behaviors and spatio-temporal computation loads of edge servers. To tackle this challenge, we transform and decompose the original task offloading optimization problem into two sub-problems, i.e., task offloading control (ToC) and server grouping (SeG). Then, we devise an online control scheme, namedMOTO(i.e.,Mobility-awareOnlineTaskOffloading), which consists of two components, i.e., Long Short Term Memory based algorithm and Dueling Double DQN based algorithm, to efficiently solve theToCandSeGsub-problems, respectively. Extensive trace-driven experiments are carried out and the results demonstrate the effectiveness ofMOTOin reducing computational costs of mobile devices and achieving load balancing when compared to the state-of-the-art benchmarks.

Synthetic traffic generation can produce sufficient data for model training of various traffic analysis tasks for IoT networks with few costs and ethical concerns. However, with the increasing functionalities of the latest smart devices, existing approaches can neither customize the traffic generation of various device functions nor generate traffic that preserves the sequentiality among packets as the real traffic. To address these limitations, this paper proposes IoTGemini, a novel framework for high-quality IoT traffic generation, which consists of a Device Modeling Module and a Traffic Generation Module. In the Device Modeling Module, we propose a method to obtain the profiles of the device functions and network behaviors, enabling IoTGemini to customize the traffic generation like using a real IoT device. In the Traffic Generation Module, we design a Packet Sequence Generative Adversarial Network (PS-GAN), which can generate synthetic traffic with high fidelity of both per-packet fields and sequential relationships. We set up a real-world IoT testbed to evaluate IoTGemini. The experiment result shows that IoTGemini can achieve great effectiveness in device modeling, high fidelity of synthetic traffic generation, and remarkable usability to downstream tasks on different traffic datasets and downstream traffic analysis tasks.

Federated learning (FL) has gained increasing attention due to its ability to collaboratively train while protecting client data privacy. However, vanilla FL cannot adapt to client heterogeneity, leading to a degradation in training efficiency due to stragglers, and is still vulnerable to privacy leakage. To address these issues, this paper proposes RingSFL, a novel distributed learning scheme that integrates FL with a model split mechanism to adapt to client heterogeneity while maintaining data privacy. In RingSFL, all clients form a ring topology. For each client, instead of training the model locally, the model is split and trained among all clients along the ring through a pre-defined direction. By properly setting the propagation lengths of heterogeneous clients, the straggler effect is mitigated, and the training efficiency of the system is significantly enhanced. Additionally, since the local models are blended, it is less likely for an eavesdropper to obtain the complete model and recover the raw data, thus improving data privacy. The experimental results on both simulation and prototype systems show that RingSFL can achieve better convergence performance than benchmark methods on independently identically distributed (IID) and non-IID datasets, while effectively preventing eavesdroppers from recovering training data.

Integrating the advantages of both the edge and the cloud, the edge-cloud computing system emerges to provide high-quality computing services for mobile users. To improve system efficiency, we investigate a hybrid mode of resource collaboration and expansion for the edge-cloud computing system, in which edge servers not only can collaborate with the cloud by purchasing high-priority computation resources temporarily but also can expand their local computation resources permanently. In such a way, the edge server can maximize its long-term profit by making a trade-off between the purchasing cost and the expanding cost. By formulating the resource management problem as a long-term profit maximization one, we first analyze the relationships among the expected minimal purchasing cost, the computation delay, and the available computation resources. Then, we design an efficient resource reserving and expanding scheme to determine the optimal expected amounts of reserving resources and expansion resources. Next, we propose an efficient real-time resource purchasing scheme to obtain the optimal amount of real-time purchasing resources dynamically. Finally, simulation results show that the proposed efficient resource collaboration and expanding scheme can maximize the long-term profit while guaranteeing the computation delay.

Cellular Internet card (IC) as a new business model emerges, which penetrates rapidly and holds the potential to foster a great business market. However, with the explosive growth of IC users, the user churn problem becomes severe, affecting the IC business significantly, while there is lacking appropriate techniques in the literature to deal with the issue. In this article, we take the lead to study one large-scale data set from a provincial network operator of China, which contains about 4 million IC users and 22 million traditional card (TC) users. We first justify the IC user churn issue with data, and categorize the user churning reasons. Then, we shed light on understanding user portraits, which is the building block to enable efficient model design. Particularly, we conduct a systematical analytics on usage data by studying the difference of two types of users, examining the impact of user properties, and characterizing the user Internet using behaviors. Finally, by using the IC user portraits and usage patterns, we propose anICuserChurnPrediction model, namedICCP, which consists of a feature extraction component and a learning-based churn prediction architecture design. For feature extraction, both the static portrait features and temporal sequential features are captured. In the learning architecture, we devise the principal component analysis (PCA) block and the embedding/transformer layers to learn the respective information of two types of features, which are collectively fed into the classification multilayer perceptron layer (MPL) for churn prediction. A reference implementation ofICCPis conducted within the telecom system and extensive experiments corroborate the efficiency ofICCP.

Federated learning (FL) enables collaborative model training over distributed computing nodes without sharing their privacy-sensitive raw data. However, in FL, iterative exchanges of model updates between distributed nodes and the cloud server can result in significant communication cost, especially when the data distributions at distributed nodes are imbalanced with requiring more rounds of iterations. In this paper, with our in-depth empirical studies, we disclose that extensive cloud aggregations can be avoided without compromising the learning accuracy if frequent aggregations can be enabled at edge network. To this end, we shed light on the hierarchical federated learning (HFL) framework, where a subset of distributed nodes can play as edge aggregators to support edge aggregations. Under the HFL framework, we formulate a communication cost minimization (CCM) problem to minimize the total communication cost required for model learning with a target accuracy by making decisions on edge aggragator selection and node-edge associations. Inspired by our data-driven insights that the potential of HFL lies in the data distribution at edge aggregators, we propose ShapeFL, i.e., SHaping dAta distRibution at Edge, to transform and solve the CCM problem. In ShapeFL, we divide the original problem into two sub-problems to minimize the per-round communication cost and maximize the data distribution diversity of edge aggregator data, respectively, and devise two light-weight algorithms to solve them accordingly. Extensive experiments are carried out based on several opened datasets and real-world network topologies, and the results demonstrate the efficacy of ShapeFL in terms of both learning accuracy and communication efficiency.

In this article, we investigate the edge cache deployment for high definition (HD) urban map provisioning, which is an essential building block for future autonomous driving. Given a deployment budget, we first formulate a satisfied downloading request maximization (SDRM) problem to obtain the deployment locations and customized cache sizes. The SDRM problem is unsolvable directly as the urban transportation traffic is highly dynamic and future traffic conditions are unknown in advance. Based on data analytics of the vehicular GPS trace, we propose an architecture named CoDe , built on which we transform and address the SDRM problem. The reference implementation of CoDe is respectively developed based on the insights from two urban-scale carpool GPS traces. The novelty and contributions of CoDe lie in its three-layer design. Particularly, at data feeding layer, we make use of two urban 60-day GPS traces involving respectively 37,801 and 17,517 vehicles, to extract the analytics samples. At mobility characterization layer, we conduct extensive data analytics on traffic mobility in terms of their distribution, correlation, and variation, to mine the crucial traffic mobility patterns for strategy customization. At cache deployment layer, we propose the K -order subgraph for each block to record the moving statistics within its K -order neighborhood, and transform the SDRM problem accordingly. Then, the R oute we I ght ba S ed gr E edy ( RISE ) algorithm is devised for the problem, which can deliver the deployment decisions. Extensive data-driven experiments are carried out to demonstrate the superior performance of CoDe in terms of request hit ratio and caching resource utility.

In distributed IoT data systems, full-size data collection is impractical due to the energy constraints and large system scales. Our previous work has investigated the advantages of integrating matrix sampling and inference for compact distributed IoT data collection, to minimize the data collection cost while guaranteeing the data benefits. This paper further advances the technology by boosting fast and accurate inference for those distributed IoT data systems that are sensitive to computation time, training stability, and inference accuracy. Particularly, we proposeCODE$^{+}$+, i.e.,Compact Distributed IOTData CollEction Plus, which features a cluster-based sampling module and a Convolutional Neural Network (CNN)-Transformer Autoencoders-based inference module, to reduce cost and guarantee the data benefits. The sampling component employs a cluster-based matrix sampling approach, in which data clustering is first conducted and then a two-step sampling is performed in accordance with the number of clusters and clustering errors. The inference component integrates a CNN-Transformer Autoencoders-based matrix inference model to estimate the full-size spatio-temporal data matrix, which consists of a CNN-Transformer encoder that extracts the underlying features from the sampled data matrix and a lightweight decoder that maps the learned latent features back to the original full-size data matrix. We implementCODE$^{+}$+under three operational large-scale IoT systems and one synthetic Gaussian distribution dataset, and extensive experiments are provided to demonstrate its efficiency and robustness. With a 20% sampling ratio,CODE$^{+}$+achieves an average data reconstruction accuracy of 94% across four datasets, outperforming our previous version of 87% and state-of-the-art baseline of 71%.

In the classical vertical federated learning (VFL) framework, feature maps and corresponding gradient information of all samples are transferred between the server and clients, which causes a significant communication burden. Therefore, to enable efficient VFL in resource-constrained wireless networks, we propose to select a part of the samples from the large training set to train models with minimal accuracy degradation. To this end, we propose LEARN, i.e., seLecting Efficient sAmples without tRaining verificatioN, to select efficient training samples for VFL. Particularly, LEARN integrates two major components named label distribution smoothing and feature center-based vertical sample filtering. The number of samples selected for each class is determined by the label distribution smoothing mechanism. Then the feature center-based vertical sample filtering component calculates the features centers and performs sample selection based on the distance between the samples and their corresponding feature center. Extensive experiments under various settings are carried out to corroborate the efficacy and robustness of LEARN.

The increasing network traffic and dynamic user connections have posed challenges for cellular operators in reducing operating costs while ensuring the quality of service (QoS) for users. Cloud radio access network (C-RAN) addresses these issues by separating baseband units (BBUs) and remote radio heads (RRHs), creating a centralized BBU pool. To optimize C-RAN performance, the key is to dynamically assigning RRHs to BBUs, which is challenging due to cost and QoS constraints. In this paper, we propose a data-driven RRH-BBU mapping scheme (KC-A3C) with deep reinforcement learning (DRL) to improve the performance of large-scale C-RANs. First, we analyze a dataset from a cellular operator containing approximately 26,652 active base stations and use the features of the dataset to construct an RRH popularity metric to cluster RRHs. Second, we model the RRH-BBU mapping as a Markov decision process and use the synchronous Advantage Actor-Critic (A3C) algorithm to find the optimal mapping scheme with the highest long-term gain in a dynamic environment, considering resource utilization, RRH migration, and BBU load balancing. Evaluations using real-world datasets show that our proposed scheme outperforms baseline methods.

Federated learning (FL) is an emerging distributed machine learning paradigm, which has shown great potential in collaborative learning with privacy preservation. However, FL clients usually have disparate system resource capabilities (e.g., data, computation, and communication) for model training and aggregation, which can cause a series of system heterogeneity issues with performance degradation. To this end, we propose FedPKD, a Prototype-based Knowledge Distillation framework for FL. FedPKD integrates knowledge distillation and prototype learning with FL, which enables heterogeneous clients and the server to learn collaboratively, with different model architectures and resource capability adaptations. Specifically, FedPKD proposes to transfer dual knowledge of clients including the model output logits and prototypes to the server, and a prototype-based ensemble distillation mechanism is proposed to aggregate the logits and prototypes from clients, which can be used to train the server model with an unlabeled public dataset. The server model knowledge is then transferred back to clients to improve the performance of client models. Moreover, to improve learning performance and reduce communication overhead, we propose a prototype-based data filter mechanism to filter out the samples with low-quality knowledge. Extensive experiments under various settings demonstrate the superiority of FedPKD in learning performance and communication efficiency when compared to state-of-the-art benchmarks.

Federated learning (FL) enables distributed clients to collaboratively learn a shared model while keeping their raw data private. To mitigate the system heterogeneity issues of FL and overcome the resource constraints of clients, we investigate a novel paradigm in which heterogeneous clients learn uniquely designed models with different architectures, and transfer knowledge to the server to train a larger server model that in turn helps to enhance client models. For efficient knowledge transfer between client models and server model, we propose FedHKT, a Hierarchical Knowledge Transfer framework for FL. The main idea of FedHKT is to allow clients with similar data distributions to collaboratively learn to specialize in certain classes, then the specialized knowledge of clients is aggregated to a super knowledge covering all specialties to train the server model, and finally the server model knowledge is distilled to client models. Specifically, we tailor a hybrid knowledge transfer mechanism for FedHKT, where the model parameters based and knowledge distillation (KD) based methods are respectively used for client-edge and edge-cloud knowledge transfer, which can harness the pros and evade the cons of these two approaches in learning performance and resource efficiency. Besides, to efficiently aggregate knowledge for conducive server model training, we propose a weighted ensemble distillation scheme with server-assisted knowledge selection, which aggregates knowledge by its prediction confidence, selects qualified knowledge during server model training, and uses selected knowledge to help improve client models. Extensive experiments demonstrate the superior performance of FedHKT compared to state-of-the-art baselines.

Motivated by many potential applications that could be enabled by acoustic motion tracking, in this paper we systematically examine the factors that limit the accuracy of acoustic tracking in practical scenarios. We identify three main challenges: (i) high mobility, (ii) low SNR, and (iii) hardware frequency response. We further show that the last two issues may exacerbate the performance issue under high mobility. We develop effective approaches to address the issues. In particular, to address high mobility, we tackle phase wrap-around using the derivative of the phase; we further estimate the Doppler shift under diverse scenarios and compensate the Doppler in channel impulse response (CIR). To address low SNR, we use a novel approach to estimate the phase shift between consecutive time intervals to effectively support time-domain beamforming and increase SNR. To tackle the uneven frequency response, we show that it is important to estimate and compensate the phase as well as the amplitude of the frequency response. Our extensive evaluation shows that each of our techniques is effective and putting them together significantly enhances the accuracy of acoustic motion tracking in general scenarios.

Mobile-edge computing (MEC) is a promising approach to support high-quality time-sensitive applications. With the increasing number of mobile devices, achieving efficient service migration management has become nontrivial in MEC. In addition, the service migration issue is difficult to be solved in real time due to user mobility and dynamic network conditions. In this article, we investigate the mobility-aware service migration problem in MEC by introducing a data-driven framework. First, service migration is formulated as an optimization problem for minimizing the long-term system delay that consists of computing, communication, and migration delays. Second, we propose a Mobility-aware Service Migration scheme, named MSM, consisting of three layers: 1) the data collection layer; 2) the association patterns analysis layer; and 3) the service migration layer. Specifically, we first collect users’ historical Wi-Fi traces to mine the association patterns. We then design a user management mechanism to reduce the complexity of decision making by using user association patterns. Finally, we formulate the service migration as a 2-D-Markov decision process and devise a deep reinforcement learning (DRL)-based algorithm to obtain service migration decisions in a large-scale MEC scenario. Extensive data-driven experiments are conducted to demonstrate the efficacy of MSM in reducing the system delay.

Map matching is a fundamental component for location-based services (LBSs), such as vehicle mobility analysis, navigation services, traffic scheduling, etc. In this paper, we investigate federated learning augmented map matching based on heterogeneous cellular moving trajectories from different operator systems, the goal of which is to improve matching accuracy without violating the user privacy. First, we develop a data collection platform with one Android-based application, and conduct rigorous data collection campaigns. Second, we perform systematic data analytics to reveal the data-driven technical challenges, including the impact of sampling rate, high location error of cellular moving data, and poor heterogeneous matching performance. Third, we propose an augmented map matching model, named FL-AMM, i.e.,FederatedLearningAugmentedMapMatching, in which we i) adopt the vertical federated learning framework to achieve data collaboration and privacy protection for heterogeneous operators; ii) devise a data augmentation component to enhance the capability of representing the raw cellular data; and iii) design a map matching model to further learn the mapping function from cellular trajectory points to road segments. Finally, we conduct extensive data-driven experiments to corroborate the efficiency and robustness of the proposed FL-AMM.

For a nationwide logistics transportation system, it is critical to make the vehicle loading plans (i.e., given many packages, deciding vehicle types and numbers) at each sorting and distribution center. This task is currently completed by dispatchers at each center in many logistics companies and consumes a lot of workloads for dispatchers. Existing works formulate such an issue as a cargo loading problem and solve it by combinatorial optimization methods. However, it cannot work in some real-world nationwide applications due to the lack of accurate cargo volume information and effective model design under complicated impact factors as well as temporal correlation. In this paper, we explore a new opportunity to utilize large-scale route and human behavior data (i.e., dispatchers' decision process on planning vehicles) to generate vehicle loading plans (i.e., plans). Specifically, we collect a five-month nationwide operational dataset from JD Logistics in China and comprehensively analyze human behaviors. Based on the data-driven analytics insights, we design a Vehicle Loading Plan learning model, named VeLP, which consists of a pattern mining module and a deep temporal cross neural network, to learn the human behaviors on regular and irregular routes, respectively. Extensive experiments demonstrate the superiority of VeLP, which achieves performance improvement by 35.8% and 50% for trunk and branch routes compared with baselines, respectively. Besides, we deployed VeLP in JDL and applied it in about 400 routes, reducing the time by approximately 20% in creating plans. It saves significant human workload and improves operational efficiency for the logistics company.

To improve the transportation efficiency of advanced manufacturing, cameras have been extensively deployed to enhance the on-road perception of autonomous ground vehicles in smart industrial parks. Considering the informative yet substantial data volume of contents generated by those cameras, we employ vehicle-to-everything links to deliver the captured video frames to neighboring vehicles, road-side units, or base stations, in order to respond to vehicle-control-related video queries. To help vehicles obtain low-latency and high-accuracy on-road information for autonomous driving, an optimization problem is formulated, taking into account the impact of vehicle mobility and diverse resource demands of different video queries. Then, a two-stage algorithm is proposed to determine the frame rate of video cameras, as well as the destination of the associated video frames based on matching theory. Extensive simulation results show that this approach can improve the accuracy by up to 18% and reduce the response delay by 13.2%.

User authentication nowadays has become an important support for not only security guarantees but also emerging novel applications. Although WiFi signal-based user authentication has achieved initial success, it works in single-user scenarios while multi-user authentication remains a challenging task. In this paper, we present MultiAuth, a multi-user authentication system that can authenticate multiple users with a single pair of commodity WiFi devices. The basic idea is to profile multipath components of WiFi signals, and leverage the multipath components to characterize each user individually for multi-user authentication. MultiAuth first profiles multipath components of WiFi signals through a proposed MUltipath Time-of-Arrival estimation algorithm (MUTA). Then, after matching corresponding multipath components to each user in complex multi-user scenarios, MultiAuth constructs individual CSI based on the multipath components to characterize each user individually. An AoA-based approach is exploited to further separate individual CSI constructed by the users with same ToA. To identify users through their activities, MultiAuth extracts user behavior profiles based on the individual CSI, and leverages a dual-task neural network for robust user authentication. Extensive experiments involving 3 simultaneously present users demonstrate that MultiAuth is effective in multi-user authentication with 86.2% average accuracy and 9.5% average false accept rate.

Vehicle platooning is an effective way to improve the efficiency and safety of transportation systems, in which a group of vehicles maintains a moving pattern by minimizing the tracking error of each vehicle. In this paper, a joint optimization of radio resource allocation for kinetic status information transmission and platoon control is considered under resource-constrained conditions to maintain the targeted inter-vehicle spacing. The formulated problem is approximately solved by the decomposition method, where the radio resource allocation and the platoon control are considered alternatively in two stages. In the first stage, a tracking error based scheduling strategy is presented for radio resource allocation. In the second stage, the control inputs of each vehicle are optimized based on the model predictive control (MPC). Simulation results show that the proposed scheme can achieve the objective of platoon control while having a low tracking error compared with other scheduling strategies.

Mobile edge computing (MEC) is a promising computing paradigm enabling mobile devices to offload computation-intensive tasks to nearby edge servers for fast processing. In this paper, we investigate the computing task offloading in small-cell MEC systems. Considering the unevenly distributed mobile users, it is critical to balance the computing load among edge servers to better utilize the computing resources. To this end, we formulate a joint task offloading control and load balancing problem to minimize the average computational cost of users. The formulated problem is a mixed-integer nonlinear optimization problem and is intractable with system scale. To solve the problem in real time, we propose a reinforcement learning-based grouping and task offloading control (RLGTC) scheme. Specifically, we first decompose the problem into two sub-problems with the Tammer method, i.e., the task offloading control (ToC) and server grouping (SeG) sub-problems. Then, we devise two algorithms based on the Kalman Filter technique and reinforcement learning with Dueling Double DQN to solve them, respectively. Extensive data-driven experiments demonstrate the effectiveness of the RLGTC scheme in achieving load balancing and reducing UEs’ computational costs compared to the state-of-the-art benchmarks.

As 360-degree videos are with high data volume, it is a great challenge to deliver the content and provide a high quality of experience (QoE) for users. In this paper, we investigate the tile-level rate allocation problem with the purpose of optimizing users’ QoE. Specifically, we find the nonlinearity between video quality and bitrate through extensive experiments. Thus, a QoE metric is defined to better measure the perceptual quality. Considering the sequential decision nature of video streaming, we formulate the rate decision problem as an Markov Decision Process. Then we propose a deep reinforcement learning based rate adaptive streaming approach to solve this problem. However, the solution space is large as a 360-degree video is spatially partitioned into multiple tiles. In order to address the problem of combinatorial explosion, we propose a tile classification method based on the predicted viewpoint. Experimental results based on real-world traces show that our algorithm can improve the overall QoE by 16% − 22% compared to existing algorithms.

Mobile Edge Computing (MEC) is a promising paradigm to support high-quality time-sensitive applications. In this paper, we investigate the service migration (i.e., whether, when, and where to migrate the services) to seamlessly serve mobile users in small-cell MEC systems. The service migration is formulated as an optimization problem to minimize the long-term system average delay that consists of queuing, communication, and migration delays. Considering the dynamic user mobility and network conditions, the formulated problem is non-convex and difficult to solve in real time. To this end, we propose a Mobility-aware Service Migration scheme, named MSM, to make real-time decisions on service migrations by utilizing reinforcement learning (RL) approaches. Specifically, we first design a user classification mechanism based on users’ mobility patterns to reduce the complexity of decision-making. We then formulate the service migration as a Markov decision process and devise an RL-based framework to make service migration decisions in real time in the dynamic MEC environment. Extensive data-driven experiments demonstrate the efficacy of MSM in reducing the system average delay.

It is unpractical to conduct full-size data collection in ubiquitous IoT data systems due to the energy constraints of IoT sensors and large system scales. Although sparse sensing technologies have been proposed to infer missing data based on partial sampled data, they usually focus on data inference while neglecting the sampling process, restraining the inference efficiency. In addition, their inferring methods highly depend on data linearity correlations, which become less effective when data are not linearly correlated. In this paper, we propose, Compact IOT Data CollEction, namely CODE, to conduct precise data matrix sampling and efficient inference. Particularly, CODE integrates two major components, i.e., cluster-based matrix sampling and Generative Adversarial Networks (GAN)-based matrix inference, to reduce the data collection cost and guarantee the data benefits, respectively. In the sampling component, a cluster-based sampling approach is devised, in which data clustering is first conducted and then a two-step sampling is performed in accordance with the number of clusters and clustering errors. For the inference component, a GAN-based model is developed to estimate the full matrix, which consists of a generator network that learns to generate a fake matrix, and a discriminator network that learns to discriminate the fake matrix from the real one. A reference implementation of CODE is conducted under three operational large-scale IoT systems, and extensive data-driven experiment results are provided to demonstrate its efficiency and robustness.

With the development of smart indoor environments, user authentication becomes an essential mechanism to support various secure accesses. Although recent studies have shown initial success on authenticating users with human activities or gestures using WiFi, they rely on predefined body gestures and perform poorly when meeting undefined body gestures. This work aims to enable WiFi-based user authentication with undefined body gestures rather than only predefined body gestures, i.e., realizing a gesture-independent user authentication. In this paper, we first explore physiological characteristics underlying body gestures, and find that statistical distributions under WiFi signals induced by body gestures can exhibit invariant individual uniqueness unrelated to specific body gestures. Inspired by this observation, we propose a user authentication system, which utilizes WiFi signals to identify individuals in a gesture-independent manner. Specifically, we design an adversarial learning-based model, which suppresses specific gesture characteristics, and extracts invariant individual uniqueness unrelated to specific body gestures, to authenticate users in a gesture-independent manner. Extensive experiments in indoor environments show that the proposed system is feasible and effective in gesture-independent user authentication.

With the explosive growth of mobile and Internet of Things (IoT) applications, increasing Mobile Edge Computing (MEC) systems have been developed by diverse Edge Service Providers (ESPs), opening a new computing market with stiff competition. However, considering the spatiotemporally varying features of computation tasks, taking over all the received tasks alone may greatly degrade the service performance of the MEC system and lead to poor economical benefit. To this end, this paper proposes a two-layer collaboration model for ESPs. Each ESP can balance the computation workload among the internal edge nodes from the ESP and offload part of computation tasks to the ESP external edge servers from other ESPs. For internal load balancing, we propose a task balancing scheme based on the Alternating Direction Method of Multipliers (ADMM) to manage the computation tasks within the edge nodes of the ESP, such that the computation delay can be minimized. For external task offloading, we formulate a game-based pricing and task allocation scheme to derive the best game strategy, aiming at maximizing the total revenue of each ESP. Extensive simulation results demonstrate that the proposed schemes can achieve improved performance in terms of system revenue and stability, as well as computation delay.

Internet card (IC) as a new business model emerges, which penetrates rapidly and holds the potential to foster a great business market. However, the understanding of IC user portraits is insufficient, which is the building block to boost the IC business. In this paper, we take the lead to bridge the gap by studying one large-scale dataset collected from a provincial network operator of China, which contains about 4 million IC users and 22 million traditional card (TC) users. Particularly, we first conduct a systematical analysis on usage data by investigating the difference of two types of users, examining the impact of user properties, and characterizing the spatio-temporal networking patterns. After that, we shed light on one specific business case of churn prediction by devising an IC user Churn Prediction model, named ICCP, which consists of a feature extraction component and a learning architecture design. In ICCP, both the static portrait features and temporal sequential features are extracted, and one principal component analysis block and the embedding/transformer layers are devised to learn the respective information of two types of features, which are collectively fed into the classification multilayer perceptron layer for prediction. Extensive experiments corroborate the efficacy of ICCP.

Edge computing servers (ECSs) have been widely deployed in large-scale mobile edge computing (MEC) systems, which can provide nearby computing services by charging users a price. Service pricing schemes can regulate user task offloading and affect the total revenue of service providers. Investigating how to maximize the revenue of service provider and improve the utilization of edge computing resources becomes crucial while is challenging, considering the users mobility and the uncertainty of users service requests. In this paper, we model the dynamic pricing process of ECS as a Markov decision process and propose a dynamic pricing approach based on Dueling Double Deep Q Network (D3QN) by using the current load conditions and user characteristics, the goal of which is to maximize the revenue of service provider. In addition, considering more ECSs in the MEC system, with the dynamic variations of ECSs loads and the different arrival rate of user tasks, we propose a joint scheduling approach based on D3QN (called RLJS) to collectively improve the total service revenue of service providers. Specifically, we first use a data-driven method to group the ECSs and then devise a D3QN-based task scheduling scheme to distribute tasks among ECS groups by considering the load and price conditions in real time. Simulation results demonstrate the efficacy of RLJS in improving the total revenue of the system provider and reducing the user delays.

Federated learning (FL) enables distributed mobile devices to collaboratively learn a shared model without exposing their raw data. However, heterogeneous devices usually have limited and different available resources, i.e., system heterogeneity, for model training and communicating, while the diverse data distribution among devices, i.e., data heterogeneity, may result in significant performance degradation. In this paper, we propose TailorFL, a dual-personalized FL framework, which tailors a submodel for each device with personalized structure for training and personalized parameters for local inference. To achieve this, we first excavate the personalization principle for data heterogeneous FL via in-depth empirical studies, and based on which, we propose a resource-aware and data-directed pruning strategy that makes each device's submodel structure match its resource capability and correlate with its local data distribution. To aggregate the submodels while preserving their dual personalization properties, we design a scaling-based aggregation strategy that scales parameters with the pruning rate of submodels and aggregates the overlapped parameters. Moreover, to further promote beneficial and restrain detrimental collaborations among devices, we propose a server-assisted model-tuning mechanism, which dynamically tunes device's submodel structure at the server side with the global view of device's data distribution similarities. Extensive experiments demonstrate that compared to the status quo approaches, TailorFL achieves an average of 22% increase in inference accuracy, and reduces the memory, computation, and communication costs for model training simultaneously.

Real-time search is essential for accessing specific sensing data (SD) in vehicular networks to support safe, efficient, and intelligent road services. Considering the tremendous data volume, the SD search process should be carefully devised to avoid excessive retrieval and transmission delay. To alleviate the communication and computational burden for sensing devices and the cloud server, roadside edges are adopted to cache the SD in advance. Given a short lifetime of SD, the caching scheme is required to be efficient in facilitating both the search process and uplink/downlink transmission, which is quite challenging due to the coupling of resource allocation decisions. To guarantee the search efficacy and enhance the caching resource utilization, we propose a real-time search-driven caching (RSC) paradigm to enable the cooperation among storage-constrained edges. A hierarchical indexing framework is first introduced for cached data, based on which we then devise a search utility model to quantify the expected data freshness and response delay. With the objective of maximizing the long-term search reward, the RSC problem is formulated by jointly considering the search requests and utility model. A deep-reinforcement-learning-based caching (DRLC) method is proposed to solve the problem. Specifically, an action transition module is introduced to lower the computational complexity via reducing the selection space of caching actions. Extensive simulations are carried out based on the real trace data in Creteil, France, and results show that the intelligent DRLC method can improve the real-time search performance significantly comparing to the benchmark methods.

In the autonomous driving era, high-definition (HD) maps are an essential building block to enable fine-grained environmental perception, precise localization, and path planning. However, with rich multidimensional information, the size of HD map data is huge and cannot be stored onboard, where the dynamic map data need to be distributed in real time via vehicular networks and how to design the distribution mechanism (i.e., determining the data source for requests) becomes crucial. For the end-to-end communication protocols (i.e., TCP/IP), the main limitation is the vehicle mobility and high dynamic of the network topology, which can degrade the transmission performance dramatically. Therefore, in this article, we propose a reinforcement learning-based data source selection scheme, named RLSS, for efficient HD map distribution in vehicular named data networking (NDN) scenarios, which aims at seeking the best data source (roadside infrastructures or nearby vehicles) in accordance with the map data requests. Specifically, in RLSS, we adopt a deep reinforcement learning-based architecture to learn a neural network as an agent to make the decision of data source selection, which can work online after offline training based on historical selection action performance. In addition, to solve the “cold start” problem for a new vehicle, we propose a model aggregation algorithm and weight update approach to learn the model parameters from its nearby vehicles, which can guarantee the performance while saving the communication cost. Finally, we implement RLSS in NS-3 by adopting the tools of the ndnSIM, SUMO, and Gym. Extensive simulations demonstrate that RLSS can significantly improve the transmission performance in terms of delay, throughput, and packet loss when compared with state-of-the-art data source selection schemes.

In this paper, we study Space-Air-Ground integrated Vehicular Network (SAGVN), and propose an online control framework to dynamically slice the SAG spectrum resource for isolated vehicular services provisioning. In particular, at a given time slot, the system makes online decisions on the request admission and scheduling, UAV dispatching, and resource slicing for different services. To characterize the impact of those parameters, we construct a time-averaged queue stability criteria by taking queue backlogs of all services into consideration, and formulate a system revenue function which incorporates the time-averaged system throughput and UAV dispatching cost. The objective is to maximize the system revenue while stabilizing the time-averaged queue, which falls into the scope of Lyapunov optimization theory. By bounding the drift-plus-penalty, the original problem can be decoupled into four independent subproblems, each of which is readily solved. The merits of our control framework are three-fold: 1) the system is able to admit and process as many requests as possible (i.e., maximizing the time-averaged throughput); 2) the time-averaged UAV dispatching cost is minimized; and 3) service queues are stabilized in the long-term. Extensive simulations are carried out, and the results demonstrate that the control framework can effectively achieve the system revenue maximization and queueing stabilization. Moreover, it can balance the trade-off among system throughput, UAV dispatching cost, and queueing states via parameter tuning. Compared with the fixed slicing, our dynamic slicing can react to the vehicular environment rapidly and achieve an average 26% of throughput improvement.

In this paper, we propose a two-level adaptive resource allocation (TARA) framework to support vehicular safety message transmissions. In particular, three types of safety messages are considered in urban vehicular networks, i.e., event-triggered messages for urgent condition warnings, periodic messages for vehicular status notifications, and messages for environmental perception. Roadside units are deployed for network management, and thus messages can be transmitted through either vehicle-to-infrastructure or vehicle-to-vehicle connections. To satisfy the requirements of different message transmissions, TARA framework consists of a group-level resource reservation module and a vehicle-level resource allocation module. Particularly, the resource reservation module is designed to allocate resources to support different types of message transmissions for each vehicle group at the first level. To learn the implicit relationship between the resource demand and message transmission requests, a supervised learning model is devised in the resource reservation module, where to obtain the training data we further propose a sequential resource allocation (SRA) scheme. Based on historical network information, SRA scheme offline optimizes the allocation of sensing resources, i.e., choosing vehicles to provide perception data, and communication resources. With resources reserved for each group, the vehicle-level resource allocation module is then devised to distribute specific resources for each vehicle to satisfy the differential requirements in real-time. Extensive simulation results demonstrate the effectiveness of TARA framework in terms of the high packet delivery ratio and low latency for message transmissions, and the high quality of collective environmental perception.

The provincial highway transportation system supports substantial cross-city transitions of people and logistics, where the prediction tasks in terms of station/road traffic, urban transitions, and individual traveling are crucial for boosting data intelligence. However, to achieve efficient prediction model design, the predictability analytics with data is the basis, but has not been sufficiently investigated in the existing literature yet. To bridge this gap, in this paper, we study one large-scale dataset collected from one provincial highway transportation system, which contains totally 21,685,765 vehicles and 351,766,743 transaction records, and conduct a comprehensive mobility analytics on its predictable performance. We first investigate the station traffic by mining its spatio-temporal correlations, then examine the multi-type urban transition flows (i.e., people flows and logistics) by demystifying the difference and similarity between the two types of behaviors, and finally analyze the uncertainty of individual traveling behaviors in terms of the destination and arriving time. After that, in accordance with the analytical findings, we cast a case study of data-driven model design for station traffic prediction. Specifically, a novel learning model is devised, named STAR, i.e., Spatio-Temporal Attention based pRediction model, which consists of station outflow/inflow temporal embedding components and spatio-temporal attention blocks to push the limit of prediction capability. Extensive experiments corroborate the efficacy of the proposed STAR.

In this paper, we investigate unmanned aerial vehicle (UAV) assisted physical layer security in multi-beam satellite enabled vehicle communications. Particularly, the UAV is exploited as a relay to improve the secure satellite-to-vehicle link, and simultaneously serves as a jammer by deliberately generating artificial noise (AN) to confuse Eve. The satellite beamforming (BF) and UAV power allocation (PA) are jointly optimized to maximize the secrecy rate of the legitimate user within a target beam while guaranteeing the quality of service (QoS) of users within other beams. Since the problem is nonconvex, we first convert it into an equivalent two-stage problem. Then, the outer-stage problem is solved by using one-dimensional search, and the inner-stage problem is transformed to a bi-convex problem by using the semi-definite relaxation (SDR) and Charnes Cooper transformation. To solve the inner-stage bi-convex problem, we propose an iterative alternating optimization algorithm, where the optimal BF is obtained by semi-definite programming (SDP), and the optimal UAV PA is subsequently obtained by solving the reformulated fractional programming problem with an iterative Dinkelbach method. The tightness of SDR and the complexity of our proposed approach are analyzed, and extensive simulations are carried out to evaluate the effectiveness of our proposed approach.

Federated learning enables distributed model training over various computing nodes, e.g., mobile devices, where instead of sharing raw user data, computing nodes can solely commit model updates without compromising data privacy. The quality of federated learning relies on the model updates contributed by computing nodes training with their local data. However, with various factors (e.g., training data size, mislabeled data samples, skewed data distributions), the model update qualities of computing nodes can vary dramatically, while inclusively aggregating low-quality model updates can deteriorate the global model quality. To achieve efficient federated learning, in this paper, we propose a novel framework namedFAIR, i.e.,Federated leArning with qualIty awaReness. Particularly,FAIRintegrates three major components: 1) learning quality estimation: we adopt the model aggregation weight (learned in the third component) to reversely quantify the individual learning quality of nodes in a privacy-preserving manner, and leverage the historical learning records to infer the next-round learning quality; 2) quality-aware incentive mechanism: within the recruiting budget, we model a reverse auction problem to stimulate the participation of high-quality and low-cost computing nodes, and the method is proved to be truthful, individually rational, and computationally efficient; and 3) auto-weighted model aggregation: based on the gradient descent method, we devise an auto-weighted model aggregation algorithm to automatically learn the optimal aggregation weights to further enhance the global model quality. Based on real-world datasets and learning tasks, extensive experiments are conducted to demonstrate the efficacy ofFAIR.

The emergency of federated learning (FL) enables distributed data owners to collaboratively build a global model without sharing their raw data, which creates a new business chance for building data market. However, in practical FL scenarios, the hardware conditions and data resources of the participant clients can vary significantly, leading to different positive/negative effects on the FL performance, where the client selection problem becomes crucial. To this end, we proposeAUCTION, anAutomated and qUality-awareClient selecTIONframework for efficient FL, which can evaluate the learning quality of clients and select them automatically with quality-awareness for a given FL task within a limited budget. To designAUCTION, multiple factors such as data size, data quality, and learning budget that can affect the learning performance should be properly balanced. It is nontrivial since their impacts on the FL model are intricate and unquantifiable. Therefore,AUCTIONis designed to encode the client selection policy into a neural network and employ reinforcement learning to automatically learn client selection policies based on the observed client status and feedback rewards quantified by the federated learning performance. In particular, the policy network is built upon an encoder-decoder deep neural network with an attention mechanism, which can adapt to dynamic changes of the number of candidate clients and make sequential client selection actions to reduce the learning space significantly. Extensive experiments are carried out based on real-world datasets and well-known learning models to demonstrate the efficiency, robustness, and scalability ofAUCTION.

With the help of non-orthogonal multiple access (NOMA), the number of connections of the beamspace multiple-input multiple-output (MIMO) systems can be improved with enhanced sum-rate performance, which constitutes beamspace MIMO-NOMA. Thus, most relevant papers focus on improving the system sum rate, which may inflict unbearable rate loss to weak users. To ensure the achievable rates of weak users, we maximize and analyze the minimal rate of the system in the single-beam case as well as the multi-beam case, where two completely different phenomena are revealed. Particularly, in the single-beam case, the maximized minimal rate of the beamspace MIMO-NOMA always grows rapidly with the signal-to-noise-ratio (SNR), and is larger than that of the beamspace MIMO using orthogonal multiple access (beamspace MIMO-OMA). However, in the multi-beam case, the maximized minimal rate of the beamspace MIMO-NOMA grows slower and slower in the high-SNR region, where it is smaller than that of the beamspace MIMO-OMA. To explain this difference, it is disclosed that the intra-beam interference in the single-beam case is ofsuccessive pattern, which is proved to have no limit on the max-min rate. In contrast, the inter-beam interference in the multi-beam case is ofmutual pattern, which is proved to restrict the max-min rate to a derived upper bound.

Traditional congestion control algorithms rely on various model-based methods to improve the end-to-end (E2E) performance of packet transmission. The resulting decisions quickly become less effective amid the dynamics of network conditions. In order to perform congestion control adaptively, reinforcement learning (RL) can be adopted to continuously learn the optimal strategy from the network environment. Oftentimes, the reward of such a learning problem is a weighted sum of multiple E2E performance metrics, such as throughput, delay, and fairness. Unfortunately, those weights can be only manually tuned based on extensive experiments. To address this issue, in this paper, we design a constrained RL algorithm for congestion control named CRL-CC to adaptively tune those weights, with the objective of effectively improving the overall E2E packet transmission performance. In particular, the multi-objective optimization problem is firstly formulated as a constrained optimization problem. Then, the Lagrangian relaxation method is leveraged to transform the constrained optimization problem into a single-objective optimization problem, which is solved by designing a multi-objective reward function with Lagrangian multipliers. Extensive experiments based on OpenAI-Gym show that the proposed CRL-CC algorithm can achieve higher overall performance in various network conditions. In particular, the CRL-CC algorithm outperforms the benchmark algorithm on Pantheon by 21.7%, 27.4%, and 5.3% in throughput, delay, and fairness, respectively.

To enhance driving safety and road intelligence for connected vehicles, the transmission of safety messages is critical in vehicular networks. In this paper, we focus on urban vehicular networks with deployed roadside units, and both vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) connections can be leveraged for message transmissions. We consider three types of safety messages: periodic messages for vehicular status notification, event-driven messages for urgent situation notification, and messages to achieve collective perception. To support different safety-related services, we develop a multi-dimensional resource allocation scheme to jointly optimize the sensing resource allocation (i.e., selecting vehicles as perception data providers), the V2I/V2V transmission mode selection, and the corresponding communication resource allocation. As the decisions on sensing resource allocation and wireless resource allocation are coupled, an iterative algorithm is proposed to solve the joint optimization problem by taking the differentiated service priorities into consideration. Extensive simulation results are presented to validate the effectiveness of the proposed resource allocation scheme.

Real-time sensing data access is essential for vehicular networks to support safe, efficient, and intelligent road services. Considering the tremendous data volume, the sensing big data search process should be carefully devised to avoid excessive retrieval delay. Edge caching can effectively alleviate the traffic burden and shorten the data downloading route, where the sensing data has to be uploaded to the edge in advance. Given a short life-time of sensing data, the caching scheme is required to be efficient in facilitating both the search process and uplink/downlink transmission, which is challenging due to the coupling of resource allocation decisions. In this paper, an edge-cloud cooperative caching scheme is proposed. Specifically, to enable real-time data search, we first introduce a hierarchical indexing framework for cached data, based on which we then devise a search utility model to quantify the expected data freshness and response delay. Aiming at maximizing the search utility, a Caching-assisted Real-time Search (CRS) problem is formulated. Due to its NP-hardness, we devise a greedy-based algorithm to solve the CRS problem. Simulation results demonstrate that the proposed cooperative caching scheme can significantly improve the data freshness and cache hit ratio comparing to the benchmark schemes.

To support multifarious vehicular services with differentiated quality-of-service (QoS) requirements, space-air-ground integrated vehicular networks (SAGVNs) are envisioned as a promising solution to provide global network connectivity, enhance network flexibility, and improve network reliability. In this paper, we investigate cooperative content delivery in the SAGVN, where vehicular content requests can be simultaneously served by multiple access points (APs) in space, aerial, and terrestrial networks. In specific, a joint optimization problem of vehicle-to-AP association, bandwidth allocation, and content delivery ratio, referred to as the ABC problem, is formulated to minimize the overall content delivery delay while satisfying vehicular QoS requirements. To address the tightly-coupled optimization variables, we propose a load- and mobility-aware ABC (LMA-ABC) scheme to solve the joint optimization problem as follows. We first decompose the ABC problem to optimize the content delivery ratio. Then the impact of bandwidth allocation on the achievable delay performance is analyzed, and an effect of diminishing delay performance gain is revealed. Based on the analysis results, the LMA-ABC scheme is designed with the consideration of user fairness, load balancing, and vehicle mobility. Simulation results demonstrate that the proposed LMA-ABC scheme can significantly reduce the cooperative content delivery delay comparing to the benchmark schemes.

Federated learning (FL) can enable distributed model training over mobile nodes without sharing privacy-sensitive raw data. However, to achieve efficient FL, one significant challenge is the prohibitive communication overhead to commit model updates since frequent cloud model aggregations are usually required to reach a target accuracy, especially when the data distributions at mobile nodes are imbalanced. With pilot experiments, it is verified that frequent cloud model aggregations can be avoided without performance degradation if model aggregations can be conducted at edge. To this end, we shed light on the hierarchical federated learning (HFL) framework, where a subset of distributed nodes are selected as edge aggregators to conduct edge aggregations. Particularly, under the HFL framework, we formulate a communication cost minimization (CCM) problem to minimize the communication cost raised by edge/cloud aggregations with making decisions on edge aggregator selection and distributed node association. Inspired by the insight that the potential of HFL lies in the data distribution at edge aggregators, we propose SHARE, i.e., SHaping dAta distRibution at Edge, to transform and solve the CCM problem. In SHARE, we divide the original problem into two sub-problems to minimize the per-round communication cost and mean Kullback-Leibler divergence of edge aggregator data, and devise two light-weight algorithms to solve them, respectively. Extensive experiments under various settings are carried out to corroborate the efficacy of SHARE.

Federated learning enables distributed learning in a privacy-protected manner, but two challenging reasons can affect learning performance significantly. First, mobile users are not willing to participate in learning due to computation and energy consumption. Second, with various factors (e.g., training data size/quality), the model update quality of mobile devices can vary dramatically, inclusively aggregating low-quality model updates can deteriorate the global model quality. In this paper, we propose a novel system named FAIR, i.e., Federated leArning with qualIty awaReness. FAIR integrates three major components: 1) learning quality estimation: we leverage historical learning records to estimate the user learning quality, where the record freshness is considered and the exponential forgetting function is utilized for weight assignment; 2) quality-aware incentive mechanism: within the recruiting budget, we model a reverse auction problem to encourage the participation of high-quality learning users, and the method is proved to be truthful, individually rational, and computationally efficient; and 3) model aggregation: we devise an aggregation algorithm that integrates the model quality into aggregation and filters out non-ideal model updates, to further optimize the global learning model. Based on real-world datasets and practical learning tasks, extensive experiments are carried out to demonstrate the efficacy of FAIR.

To enable fast content delivery for delay-sensitive applications, large content providers build edge servers, Points of Presence (PoPs), and datacenters around the world. They are networked together as an integrated infrastructure via a private wide-area network (WAN), named content delivery network (CDN). To deliver quality services in the CDN, there are two critical decisions that should be properly made: 1) making assignments of PoP and datacenter for user requests, and 2) selecting routing paths from PoP to datacenter. However, with both the network variability and CDN environment complexity, it is challenging to achieve satisfying decisions. In this paper, we propose DeepDelivery, an adaptive deep reinforcement learning approach to intelligently make assignments and routing decisions in real time. Essentially, DeepDelivery adopts the Markov decision process (MDP) model to capture the dynamics of network variation, and the objective is to jointly maximize the infrastructure utilization of providers and minimize the total latency of end users. We conduct extensive trace-driven evaluations spanning various environment dynamics with both real-world and synthetic trace data. The result demonstrates that DeepDelivery can outperform the state-of-the-art scheme by 21.89% higher utilization and 11.27% lower end-to-end latency on average.

In this article, we investigate the worker (i.e., vehicle) selection problem in vehicle-based crowdsourcing (VBC), where vehicles in a specific area are recruited by the crowdsourcing platform to collect geographical information in real driving scenarios for autonomous driving. Given a limited recruitment budget, we formulate a cumulative platform utility maximization problem (CMP) to obtain the optimal worker set. The CMP is unsolvable directly as the platform has no prior information of workers at the initial stage (also known as “cold start”) and the cost of collecting all workers' information is prohibitive. To solve the problem, we first conduct a comprehensive data analytics on two real-world vehicle traces and obtain two crucial observations: 1) trajectory of individual vehicle is highly uncertain that it is difficult to make accurate prediction and 2) the overall distribution of vehicular trajectory penetration (measured by collection quantity and coverage) has a diurnal pattern and varies with weekly periodicity. Inspired by the insights, we propose the performance transfer-based online worker selection (POSE) scheme, which works independently from trajectory prediction with two components, i.e., transfer learning-based performance estimation and online worker selection (OWS). Based on the diurnal pattern, the former component collects a short-period trajectory penetration data of vehicles for model fitting, which can output a specific numerical distribution. With the fitting model, we can identify and select vehicles with high trajectory penetration at the initial stage to cope with the “cold start” problem. Then, we map the worker selection problem into a multiarmed bandit problem and develop upper confidence bound-based approach to solve it. Extensive trace-driven simulations are carried out and the results demonstrate the efficiency of POSE in terms of cumulative platform utility.

In this article, we proposeFLAGfor flexible, accurate, and long-time user load prediction in a large-scale WiFi system.FLAGenables prediction customization in both time granularity and prediction length. Under an operating WiFi system with more than 7000 APs, a reference implementation ofFLAGis developed, which consists of three major components. Fordata acquisition, we process 25 074 733 association records contributed by 55 809 users, to extract the ground truth of AP-level user load. Forfeature extraction, we perform a comprehensive data analytics to mine vital features to label each AP, which are extracted and classified into three categories, i.e., individual features, spatial features, and temporal features. For themodel design, we design a deep recurrent neural network (RNN) model, which contains two separate RNNs, i.e., the encoder RNN and decoder RNN. Particularly, the sequential feature vectors are injected into the encoder RNN to learn the “semantic” information, based on which the decoder RNN conducts sequential AP-level predictions. As the semantic vector is injected for each time step prediction, it can effectively reduce the accumulated prediction errors, which enable long period of time predictions. Real data set-based experiments corroborate the efficacy ofFLAG.

Connected vehicles have been considered as an effective solution to enhance driving safety as they can be well aware of nearby environments by exchanging safety beacons periodically. However, under dynamic traffic conditions, especially for dense-vehicle scenarios, the naive beaconing scheme where vehicles broadcast beacons at a fixed rate with a fixed transmission power can cause severe channel congestion and thus degrade the beaconing reliability. In this paper, by considering the kinematic status and beaconing rate together, we study the rear-end collision risk and define a danger coefficient ρ to capture the danger threat of each vehicle being in the rear-end collision. In specific, we propose a fully distributed adaptive beacon control scheme, called ABC, which makes each vehicle actively adopt a minimal but sufficient beaconing rate to avoid the rear-end collision in dense scenarios based on individually estimated ρ. With ABC, vehicles can broadcast at the maximum beaconing rate when the channel medium resource is enough and meanwhile keep identifying whether the channel is congested. Once a congestion event is detected, an NP-hard distributed beacon rate adaptation (DBRA) problem is solved with a greedy heuristic algorithm, in which a vehicle with a higher ρ is assigned with a higher beaconing rate while keeping the total required beaconing demand lower than the channel capacity. We prove the heuristic algorithm's close proximity to the optimal result and thoroughly analyze the communication overhead of ABC scheme. By using Simulation of Urban MObility (SUMO)-generated vehicular traces, we conduct extensive simulations to demonstrate the efficacy of our proposed ABC scheme. Simulation results show that vehicles can adapt beaconing rates according to the driving safety demand, and the beaconing reliability can be guaranteed even under high-dense vehicle scenarios.

Widespread and large-scale WiFi systems have been deployed in many corporate locations, while the backhual capacity becomes the bottleneck in providing high-rate data services to a tremendous number of WiFi users. Mobile edge caching is a promising solution to relieve backhaul pressure and deliver quality services by proactively pushing contents to access points (APs). However, how to deploy cache in large-scale WiFi system is not well studied yet quite challenging since numerous APs can have heterogeneous traffic characteristics, and future traffic conditions are unknown ahead. In this paper, given the cache storage budget, we explore the cache deployment in a large-scale WiFi system, which contains 8,000 APs and serves more than 40,000 active users, to maximize the long-term caching gain. Specifically, we first collect two-month user association records and conduct intensive spatio-temporal analytics on WiFi traffic consumption, gaining two major observations. First, per AP traffic consumption varies in a rather wide range and the proportion of AP distributes evenly within the range, indicating that the cache size should be heterogeneously allocated in accordance to the underlying traffic demands. Second, compared to a single AP, the traffic consumption of a group of APs (clustered by physical locations) is more stable, which means that the short-term traffic statistics can be used to infer the future long-term traffic conditions. We then propose our cache deployment strategy, named LeaD (i.e., Large-scale WiFi Edge cAche Deployment), in which we first cluster large-scale APs into well-sized edge nodes, then conduct the stationary testing on edge level traffic consumption and sample sufficient traffic statistics in order to precisely characterize long-term traffic conditions, and finally devise the TEG (Traffic-wEighted Greedy) algorithm to solve the long-term caching gain maximization problem. Extensive trace-driven experiments are carried out, and the results demonstrate that LeaD is able to achieve the near-optimal caching performance and can outperform other benchmark strategies significantly.

Named Data Networking (NDN) is considered as a prominent architecture towards future Wireless Local Area Networks (WLAN), and multicast plays an important role in data delivery such as media streaming, multipoint videoconferencing, etc. However, to achieve high-efficiency multicast in NDN WLAN is challenging for two significant reasons. First, without feedback mechanism in IEEE 802.11 standards, to guarantee reliability, the current multicast scheme transmits the multicast data with the basic rate (e.g., 1 Mbps for IEEE 802.11b), which inevitably increases the transmission delay for high-speed consumers. Second, as a NDN multicast group is constituted by consumers who are requesting the same content, multicast groups are easy to form and evolve rapidly, where a data rate adaptation scheme is requisite to accommodate differential multicast groups. In this paper, we propose a multi-stage multicast rate adaptation scheme for NDN WLAN, namedNDN-MMRA, to minimize the total transmission time with reliability guarantee for multicast group members. InNDN-MMRA, by checking the Pending Interest Table (PIT) status information, the number of consumers in each multicast group as well as their receiving capabilities are known ahead; with the available data rates in a specific 802.11 standard,NDN-MMRAdetermines: 1) how many transmission stages are required; and 2) in each stage, which data rate should be adopted. The merit is that with multi-stage transmissions, the data rate can be adapted in descending order to accommodate high-speed consumers with delay minimized, and low-speed consumers with reliability guaranteed. We implementNDN-MMRAin NS-3 by adopting the ndnSIM module, and conduct extensive experiments to demonstrate its efficacy under different IEEE 802.11 standards and various underlying WLAN topologies.

In Named Data Networking (NDN) architecture, data can be obtained from multiple content sources (i.e., producers or caching nodes) with multiple paths, making the traditional end-to-end (i.e., TCP/IP) congestion control scheme invalid. In addition, the NDN multi-path discovery and management are still an open issue as the dynamic network topology changes. In this article, we propose a multi-path congestion control mechanism, named MPCC, which includes two major components, i.e., multi-path discovery and multi-path congestion control. Particularly, for multi-path discovery, we first devise apath tagto uniquely mark each sub-path in the forwarding process, and then propose a tag-aware forwarding strategy to discover and manage sub-paths. For multi-path selection and congestion control, we first integrate the metrics of packet loss, bandwidth, round trip time, and path centrality, for path assessment, based on which, we then leverage the Upper Confidence Bound (UCB) algorithm to select sub-paths in order to maximize the network throughput. In addition, for selected sub-paths, we have devised a sub-path window adaptation algorithm to avoid multi-path congestions. At last, we implement MPCC in ndnSIM and conduct extensive experiments for performance evaluation. Our results demonstrate that MPCC can discover all sub-paths in real-time for the multi-path scenario, and can effectively avoid multi-path congestions with improving throughput and reducing transmission time.

In this paper, we propose a secure computationoffloading scheme (SCOS) in intelligently connected vehicle (ICV) networks, aiming to minimize overall latency of computing via offloading part of computational tasks to nearby servers in small cell base stations (SBSs), while securing the information delivered during offloading and feedback phases via physical layer security. Existing computation offloading schemes usually neglected time-varying characteristics of channels and their corresponding secrecy rates, resulting in an inappropriate task partition ratio and a large secrecy outage probability. To address these issues, we utilize an ergodic secrecy rate to determine how many tasks are offloaded to the edge, where ergodic secrecy rate represents the average secrecy rate over all realizations in a time-varying wireless channel. Adaptive wiretap code rates are proposed with a secrecy outage constraint to match time-varying wireless channels. In addition, the proposed secure beamforming and artificial noise (AN) schemes can improve the ergodic secrecy rates of uplink and downlink channels even without eavesdropper channel state information (CSI). Numerical results demonstrate that the proposed schemes have a shorter system delay than the strategies neglecting time-varying characteristics.

In this article, we investigate a computing task scheduling problem in space-air-ground integrated network (SAGIN) for delay-oriented Internet of Things (IoT) services. In the considered scenario, an unmanned aerial vehicle (UAV) collects computing tasks from IoT devices and then makes online offloading decisions, in which the tasks can be processed at the UAV or offloaded to the nearby base station or the remote satellite. Our objective is to design a task scheduling policy that minimizes offloading and computing delay of all tasks given the UAV energy capacity constraint. To this end, we first formulate the online scheduling problem as an energy-constrained Markov decision process (MDP). Then, considering the task arrival dynamics, we develop a novel deep risk-sensitive reinforcement learning algorithm. Specifically, the algorithm evaluates the risk, which measures the energy consumption that exceeds the constraint, for each state and searches the optimal parameter weighing the minimization of delay and risk while learning the optimal policy. Extensive simulation results demonstrate that the proposed algorithm can reduce the task processing delay by up to 30% compared to probabilistic configuration methods while satisfying the UAV energy capacity constraint.

Large-scale WiFi system is gaining an increasing momentum rapidly in most corporate places. Enabling edge functions on the system is imperative to support unprecedented edge applications. However, building edge functionalities at each AP may incur frequent service migrations, low resource utilization, and inflexible resource provisioning. It is thus prospective to federate suitable APs to create a resource-pooled edge system such that users in association with federated APs can share the pooled resource. In this paper, we propose a novel architecture, named SoSA, to Socialize Static APs via user association transition activities for edge resource pooling. A reference implementation of SoSA is developed under an operating large-scale WiFi system in a campus area of 3.0925 km2. The novelty and contribution of SoSA lie in its three-layer design. In transition data feeding layer, we collect and process 25,074,733 association records of 55,809 users from 7,404 APs in a real WiFi system. In sociality construction and characterization layer, we construct an AP contact graph based on user transition statistics, under which we empirically study the sociality of APs and explore their evolving patterns. In edge resource pooling layer, by harnessing the AP sociality, we are able to customize resource pooling strategy to improve service provisioning performance. With adopting SoSA, we systematically investigate the performance of AP federation strategy in reducing service migration when users frequently transit among APs. Extensive data-driven experiments corroborate the efficacy of SoSA.

In this paper, we propose an online control frame-work to dynamically slice the network resource for isolated service provisioning in Space-Air-Ground integrated Vehicular Network (SAGVN). In particular, at a given time slot, the system makes online decisions on the request admission and scheduling, UAV dispatching, and resource slicing for different services. To characterize the impact of those parameters, we construct a time-averaged queue stability criteria by taking queue backlogs of all services into consideration, and formulate a system revenue function which incorporates the time-averaged system throughput and UAV dispatching cost. The objective is to maximize the system revenue while stabilizing the time-averaged queue, which can be achieved via the Lyapunov optimization theory. By bounding the drift-plus-penalty, the problem then can be decoupled into four independent subproblems, which are readily solved. The merits of our control framework are three-fold: 1) the system can admit and process as many requests as possible; 2) the time-averaged UAV dispatching cost is minimized; and 3) service queues can be stabilized over time. Extensive simulations are carried out, and the results demonstrate that the control framework can effectively achieve the system revenue maximization and queueing stabilization. Moreover, it can balance the trade-off among system throughput, UAV dispatching cost, and queueing states via parameter tuning.

With the emergence of automated driving technology, wireless demand for secure information exchange among automated vehicles has increased dramatically. To this end, we design a security-aware dynamic device-to-device (D2D) spectrum resource sharing mechanism to enhance the security of vehicular D2D communications with the improved spectrum efficiency. Considering both resource block (RB) sharing and power control, we model this joint D2D spectrum resource sharing process as a weighted bipartite graph matching problem whose weights are obtained through deriving the closed-form solutions of power control in an algebraic method. Then, the global optimal solutions of the matching problem are obtained by using the Hungarian algorithm. Furthermore, a security-aware RB and power allocation (SA-RBPA) mechanism compatible with existing cellular networks is proposed for the small cell base station applications. Extensive simulation results have shown that, compared with existing approaches that consider RB reusing strategy or power control scheme optimization alone, the SA-RBPA scheme is able to realize a better spectrum efficiency and security performance.

In this article, we investigate the UAV-aided edge caching to assist terrestrial vehicular networks in delivering high-bandwidth content files. Aiming at maximizing the overall network throughput, we formulate a joint caching and trajectory optimization (JCTO) problem to make decisions on content placement, content delivery, and UAV trajectory simultaneously. As the decisions interact with each other and the UAV energy is limited, the formulated JCTO problem is intractable directly and timely. To this end, we propose a deep supervised learning scheme to enable intelligent edge for real-time decision-making in the highly dynamic vehicular networks. In specific, we first propose a clustering-based two-layered (CBTL) algorithm to solve the JCTO problem offline. With a given content placement strategy, we devise a time-based graph decomposition method to jointly optimize the content delivery and trajectory design, with which we then leverage the particle swarm optimization (PSO) algorithm to further optimize the content placement. We then design a deep supervised learning architecture of the convolutional neural network (CNN) to make fast decisions online. The network density and content request distribution with spatio-temporal dimensions are labeled as channeled images and input to the CNN-based model, and the results achieved by the CBTL algorithm are labeled as model outputs. With the CNN-based model, a function which maps the input network information to the output decision can be intelligently learnt to make timely inference and facilitate online decisions. We conduct extensive trace-driven experiments, and our results demonstrate both the efficiency of CBTL in solving the JCTO problem and the superior learning performance with the CNN-based model.

Internet-of-Vehicles (IoV) connects vehicles, sensors, pedestrians, mobile devices, and the Internet with advanced communication and networking technologies, which can enhance road safety, improve road traffic management, and support immerse user experience. However, the increasing number of vehicles and other IoV devices, high vehicle mobility, and diverse service requirements render the operation and management of IoV intractable. Software-defined networking (SDN) and network function virtualization (NFV) technologies offer potential solutions to achieve flexible and automated network management, global network optimization, and efficient network resource orchestration with cost-effectiveness and are envisioned as a key enabler to future IoV. In this article, we provide an overview of SDN/NFV-enabled IoV, in which SDN/NFV technologies are leveraged to enhance the performance of IoV and enable diverse IoV scenarios and applications. In particular, the IoV and SDN/NFV technologies are first introduced. Then, the state-of-the-art research works are surveyed comprehensively, which is categorized into topics according to the role that the SDN/NFV technologies play in IoV, i.e., enhancing the performance of data communication, computing, and caching, respectively. Some open research issues are discussed for future directions.

To develop smart city and intelligent manufacturing, video cameras are being increasingly deployed. In order to achieve fast and accurate response to live video queries (e.g., license plate recording and object tracking), the real-time high-volume video streams should be delivered and analyzed efficiently. In this article, we introduce an end-edge-cloud coordination framework for low-latency and accurate live video analytics. Considering the locality of video queries, edge platform is designated as the system coordinator. It accepts live video queries and configures the related end cameras to generate video frames that meet quality requirements. By taking into account the latency constraint, edge computing resources are subtly distributed to process the live video frames from different sources such that the analytic accuracy of the accepted video queries can be maximized. Since the amount of required edge computing resource and video quality to accurately address different video queries are unknown in advance, we propose an online video quality and computing resource configuration algorithm to gradually learn the optimal configuration strategy. Extensive simulation results show that as compared to other benchmarks, the proposed configuration algorithm can effectively improve the analytic accuracy, while providing low-latency response.

The IEEE 802.11p-based dedicated short range communication (DSRC) is essential to enhance driving safety and improve road efficiency by enabling rapid cooperative message exchanging. However, there is a lack of good understanding on the DSRC performance in urban environments for vehicle-to-vehicle (V2V) communications, which impedes its reliable and efficient application. In this paper, we first conduct intensive data analytics on V2V performance, based on a large amount of real-world DSRC communications trace collected in Shanghai city, and obtain several key insights as follows. First, among many context factors, the non-line-of-sight (NLoS) link condition is the major factor degrading V2V performance. Second, the durations of line-of-sight (LoS) and NLoS transmission conditions follow power law distributions, which indicate that the probability of experiencing long LoS/NLoS conditions both could be high. Third, the packet inter-reception (PIR) time distribution follows an exponential distribution in the LoS conditions but a power law in the NLoS conditions, which means that the consecutive packet reception failures rarely appear in the LoS conditions but can constantly appear in the NLoS conditions. Based on these findings, we propose a context-aware reliable beaconing scheme, called CoBe, to enhance the broadcast reliability for safety applications. The CoBe is a fully distributed scheme, in which a vehicle first detects the link condition with each of its neighbors by machine learning algorithms, then exchanges such link condition information with its neighbors, and finally selects the minimal number of helper vehicles to rebroadcast its beacons to those neighbors in bad link condition. To analyze and evaluate the CoBe performance, a two-state Markov chain is devised to model beaconing behaviors. The extensive trace-driven simulations are conducted to demonstrate the efficacy of CoBe.

Vehicle-based crowdsourcing is expected to be an economic yet efficient solution to build and maintain an accurate, fine-grained, and up-to-date environment map (i.e., high-definition map) for autonomous vehicles, which is an essential building block for safe and intelligent autonomous driving. However, how to select crowdsourcing workers with performance maximization is prudent and quite challenging since vehicles are highly dynamic and have unpredictable routes. In this paper, we study the worker selection problem for crowdsourced on-route map collection where the trade- off between the real-time worker exploration and exploitation is the main focus. Specifically, by adopting the multi-armed bandit model, we formulate a cumulative platform utility maximization problem. To solve this problem, we propose an Online Worker Selection (OWS) scheme, to learn drivers' performance and make worker selection decisions in real time. Essentially, two key designs are integrated in OWS: 1) performance transfer. If a new driver joins the crowdsourcing, we will initialize the new driver's performance based on the knowledge transferred from the existing drivers' records; and 2) marginal utility. Particularly, we carefully incorporate the platform utility to embody the marginal effect, i.e., repeated coverage by multiple vehicles on a certain road will undermine the utility. Based on the real-world vehicular GPS trace, we conduct extensive trace- driven simulations, and results demonstrate that our scheme can effectively obtain high-quality environment map, with on average 40.5% crowdsourcing utility gain over other benchmark schemes.

In this paper, we investigate the edge caching and content delivery problem for both high-speed train (HST) passengers and low-mobility cellular users. Under multi-dimensional resources constraints, we formulate an optimization problem to minimize the content retrieval delay of HST passengers and meanwhile guarantee the delay requirements of cellular users. As the formulated problem is a mixed-integer nonconvex optimization problem, which is intractable directly, we propose an efficient iterative algorithm that optimizes the three decision variables (i.e., content placement, subchannel allocation, and transmission power allocation) alternately. In specific, Lagrangian multiplier is introduced to convert the constrained optimization, which transforms the content caching problem into a Lagrangian relaxed knapsack problem. Afterwards, the subchannel assignment problem is solved by the Hungarian algorithm with polynomial time complexity, and the power allocation strategy is obtained by the bisection method. Extensive simulations are carried out and results demonstrate that our proposed caching strategy can reduce the content retrieval delay by up to 25% in comparison with the benchmark strategy.

In this paper, we study a multi-beam millimeter- wave beamspace multiple-input multiple-output (MIMO) system with non-orthogonal multiple access (NOMA) to simultaneously accommodate multiple users in a single beam. To improve the data rate while maintaining user fairness, we analyze the max-min rate of the system via power allocation. The challenge is that the existence of both the intra- beam and inter-beam interference makes the power- allocation problem non-convex. To address this issue, we devise a bisection approach to calculate the max-min rate and the corresponding power allocation. We prove that the max-min rate can be achieved when all the users are assigned the same rate. Furthermore, our endeavors reveal that beamspace MIMO-NOMA outperforms the traditional beamspace MIMO in terms of the minimal rate when the power or the number of users is relatively small. When the power or the number of users is relatively large, traditional beamspace MIMO can outperform beamspace MIMO-NOMA since the former is free of inter-beam interference, which has been verified by simulation results.

In this paper, we study the secrecy provisioning downlink transmission in an aerial-assisted network, where the unmanned aerial vehicle (UAV) serves as an aerial platform to provide secure transmission for the mobile users (MUs) with coexist of Internet of Things (IoT) nodes (INs). Specifically, secure transmission is required for MUs to combat eavesdropping attacks and a desired successful transmission probability should be ensured for INs to receive the public instruction massages. To improve the secrecy rates (SRs) for MUs, we consider an eavesdropper-free area, i.e., protected zone, surrounding the UAV. With non-orthogonal multiple access (NOMA) for MUs, the power allocation to each MU is optimized to maximize the minimum secrecy rate of MUs within the protected zone, under the constraints of successful receiving probability requirements for INs. To solve this problem, we first prove that the max-min SR can be obtained when SRs of all users are equal, and then a dichotomy-based successive power allocation policy is proposed. Numerical results show that higher max-min secrecy rate can be achieved by our proposed power allocation policy than the traditional policy.

Due to the versatile networking capability, space- air-ground integrated network (SAGIN) becomes a prominent future architecture to support the ever- increasing Internet of Things (IoT) applications. In this paper, we investigate the IoT task offloading under an SAGIN scenario where multiple IoT devices generate computing tasks to be processed. We adopt an unmanned aerial vehicle (UAV) to fly along a given trajectory to collect the tasks of IoT devices within the coverage area, and then makes the online offloading decision, i.e., processing locally, or offloading to the nearby base station or the far-away satellite. However, due to the constrained energy resources committed by UAV and the uncertainty of the system dynamics, designing an efficient computation task offloading algorithm is challenging. This dynamic scheduling problem is formulated as a constrained Markov decision process (CMDP), considering the stochastic channel conditions, UAV coverage, energy consumption, and task queue backlogs. By exploiting the stationary stochastic feature of the CMDP, the problem can be solved by the linear programming to find a stochastic policy. Simulation results demonstrate that the proposed computation offloading scheme can significantly reduce IoT task processing delay as compared to other benchmarks.

Multiuser multiple input multiple output (MU-MIMO) millimeter wave (mmWave) communication is considered as a key technology to provide multi-gigabit train-to-ground wireless connections in the high-speed railway (HSR) system. Considering the power consumption and hardware constraint, the hybrid beamforming (BF), which combines analog BF and digital BF, is widely adopted in the MU-MIMO mmWave systems. In this paper, we target on an efficient hybrid BF structure design in HSR scenario with taking the practical HSR mmWave channel model into consideration. Specifically, the hybrid BF design aims at maximizing the overall throughput and is formulated as an optimization problem which is proved to be nonconvex and NP-hard. Therefore, a suboptimal yet efficient two-stage solution is proposed, where a weighted minimum mean square error (WMMSE) based beamforming strategy is exploited to devise the hybrid beamformer at the base station (BS) at the first stage, and the orthogonal matching pursuit (OMP) approach is leveraged to decouple the digital BF and analog BF at BS at the second stage. Simulation results demonstrate that higher overall throughput can be achieved by the proposed hybrid BF scheme compared to other state-of-the-art benchmarks.

Large-scale WiFi systems have been widely deployed in an increasing number of corporate places such as universities, big malls and companies, to provide fast Internet experience to users. However, user association patterns in such large-scale systems have not been well investigated, which is crucial for performance enhancement and intelligent system management. In this paper, we provide the analytics of a large-scale campus WiFi dataset, which includes more than 8,000 access points (APs) and 40,000 active users in the area of 3.0925 km2. By conducting extensive analysis on association patterns, we achieve several key insights as follows. First, user associations are highly dynamic as short association durations and frequent AP transitions prevail throughout the whole trace. Second, even though users may associate to many APs, they generally have a small preferable AP set in which they spend most of their WiFi connection time for data traffic; in addition, each user has distinct yet relatively fixed AP transition route, indicating that given its current associated AP, its next association AP is highly predictable. Third, diurnal association patterns are observed not only at single AP level, but also at the building and the system level, where the number of associated users and the data traffic vary periodically on a daily basis. These insights can provide valuable guidelines to numerous intelligent service provisions such as proactive service migration, edge content distribution, efficient network management.

Ensuring network QoS for Internet of vehicles (IoVs) is crucial for safe and intelligent transportation system, while the vehicle density variation seems invincible for stationary base station (BS) networks. In this paper, we study IoV's downlink throughput guarantee, in which, in addition to the cellular BS resource, UAVs (equipped with WiFi interfaces) can be dynamically sent out to provide additional wireless connections. To cope with the dynamic IoV density, we propose an Online UAV Scheduling scheme, referred to as OUS, to online schedule and manage UAVs to guarantee seamless connections with reliable throughput performance. In OUS, we first use the complementary cumulative distribution function (CCDF) of IoV throughput to calculate the likelihood of a channel resource shortage. If a shortage condition is imminent and then minimal UAVs will be sent out to their optimal hovering positions. In particular, we revealed the marginal effect for the optimal hovering position acquisition, i.e., the further the UAV is away from the BS, the larger throughput gain can be achieved by the system. We conduct extensive simulations to evaluate the performance of our OUS scheme, and results demonstrate that it can well react to the throughput QoS demand by intelligently sending out minimal UAVs, and its hovering position acquisition method can fully utilize the efficacy of UAVs.

Drone cell (DC) is an emerging technique to offer flexible and cost-effective wireless connections to collect Internet-of-things (IoT) data in uncovered areas of terrestrial networks. The flying trajectory of DC significantly impacts the data collection performance. However, designing the trajectory is a challenging issue due to the complicated 3D mobility of DC, unique DC-to-ground (D2G) channel features, limited DC-to-BS (D2B) backhaul link quality, etc. In this paper, we propose a 3D DC trajectory design for the DC-assisted IoT data collection where multiple DCs periodically fly over IoT devices and relay the IoT data to the base stations (BSs). The trajectory design is formulated as a mixed integer non-linear programming (MINLP) problem to minimize the average user-to-DC (U2D) pathloss, considering the state-of-the-art practical D2G channel model. We decouple the MINLP problem into multiple quasi-convex or integer linear programming (ILP) sub-problems, which optimizes the user association, user scheduling, horizontal trajectories and DC flying altitudes of DCs, respectively. Then, a 3D multi-DC trajectory design algorithm is developed to solve the MINLP problem, in which the sub-problems are optimized iteratively through the block coordinate descent (BCD) method. Compared with the static DC deployment, the proposed trajectory design can lower the average U2D pathloss by 10-15 dB, and reduce the standard deviation of U2D pathloss by 56%, which indicates the improvements in both link quality and user fairness.

The energy consumption for wireless interface is important for the power-constraint mobile and sensor devices. To improve energy efficiency in WLAN (such as Wi-Fi), power saving mode (PSM) is proposed, with an attempt to manage the time spent in idle listening (IL) state. The challenge is that the receiver has no knowledge about when the pending data will arrival under end-to-end communication protocols (TCP/IP); therefore each station has to spend more time in IL to wait for the pending data. To address this problem, we propose NDN-PSM, in which NDN communication architecture is leveraged to cut down unnecessary IL time. In particular, we introduce two new power states in NDN-PSM, i.e., light doze and deep doze. As stations can check pending interest table (PIT) information to predict data arrival precisely, they can switch to deep doze or light doze intelligently. The inherent receiver-driven patterns of NDN can make each station effectively go to deep doze state for power saving. We have implemented NDN-PSM in NS-3 through ndnSIM and the simulation results demonstrate that NDN-PSM can effectively reduce IL time as well as total power consumption and meanwhile retain low transmission delay. Specifically, compared to the PSM mechanism, NDN-PSM can reduce the average power consumption up to 56%.

Mobile edge computing extends computing and storage resources to the proximity of mobile users, facilitating a number of innovative mobile applications. Particularly, video streaming is the most prevailing one that consumes substantial edge resources. In this paper, we investigate the multi-dimensional resource allocation for video service provisioning, with the objective of ensuring satisfied streaming experience at high resource utilization. Considering the diversified and constantly changing user preferences on the quality of video contents, the edge resource allocation process is modeled as a long-term utility maximization problem. To address this problem, we propose an online learning algorithm that actively estimates user preferences according to regression analysis on user feedback. This algorithm requires no training phase, and hence is adaptive to dynamic user interests and available edge resources. Both theoretical analysis and numerical results demonstrate that the performance of the proposed algorithm asymptotically approaches the hindsight optimal resource allocation strategy.

How to deploy cache in large-scale WiFi system is not well studied yet quite challenging since numerous Aps turn to be heterogeneous in terms of traffic consumption, and future traffic conditions are unknown ahead. In this paper, given the cache storage budge, we explore the cache deployment in a large-scale WiFi system which contains 8,000 APs and serves more than 40,000 active users, to maximize the long-term caching gain, i.e., the total reduced backhaul traffic. Specifically, we first collect enormous user association records and conduct intensive statistical analysis on the collected data, gaining two major observations. First, per AP traffic consumption varies in a rather wide range and the AP proportion distributes evenly within the range, which indicates that the cache size should be heterogeneously allocated in accordance to the underlying traffic demands. Second, compared to a single AP, the traffic consumption of a group of APs (clustered by physical locations) is more stable, which means that the short-term traffic statistics can be used to infer the future long-term traffic conditions. We then propose our cache deployment strategy, named LEAD (i.e., Large-scale wifi Edge cAche Deployment), in which we first cluster large-scale APs into well-sized edge nodes, then conduct the stationary testing on edge level traffic consumption and sample sufficient traffic statistics in order to precisely characterize future traffic conditions, and finally devise the TEG (Traffic-wEighted Greedy) algorithm to solve the long-term caching gain maximization problem. Extensive trace-driven simulations are carried out and simulation results demonstrate the efficacy of LEAD.

Spectrally efficient frequency division multiplexing (SEFDM) is a promising non-orthogonal multi-carrier technique to improve spectral efficiency, by compressing the inter-carrier interval relative to orthogonal frequency division multiplexing (OFDM) systems. However, by breaking the orthogonality among subcarriers, the self- introduced inter-carrier interference (ICI) severely restrains the achievable transmission rate and poses great challenges in designing the receiver with ICI cancellation. In this paper, we first characterize the statistical distribution of ICI and then derive a closed-form expression of the signal-to- interference-plus-noise ratio (SINR). After that, an efficient time-domain ICI mitigation approach is proposed to improve the achievable SINR and the spectral efficiency of the SEFDM system. Numerical results verify the analytical expressions for the cumulative distribution function (CDF) of ICI and the achievable SINR. In addition, it is shown that the spectral efficiency can be significantly improved by adopting our proposed ICI mitigation approach.

Full WiFi coverage becomes more and more prevalent in corporate places, such as university, big mall, airport, and so forth. To achieve full WiFi coverage in a wide area is costly due to the large-scale AP deployment spending and considerable operating expenditure. However, with limited literature available, how to deploy and manage those APs in an efficient and economical way, is still unknown for system providers. To bridge this gap, in this article, we first collect large-scale AP usage data in our campus WiFi system, which contains over 8000 APs and serves more than 40 000 active end-users in the area of 3.0925 km2. After mining large-scale spatio-temporal user associations, we obtain several key insights as follows. First, Idle Phenomenon prevails throughout the trace, in which large portion of APs are wasted without any user association. Second, AP usages in different buildings have very distinct characteristics in terms of user association and traffic consumption. Third, diurnal usage patterns are very obvious not only at singe AP level but also at the building and the whole system level. Many deployment and management strategies can benefit from these insights, e.g., heterogeneous AP deployment and intelligent AP management. Among them, we then propose an intelligent large-scale AP management scheme, called LAM, to dynamically control large-scale APs (ON or OFF) for energy saving and meanwhile without loss of WiFi coverage. In LAM, based on history association records, the user load of each AP is predicted by machine learning algorithms, and those APs whose idle durations are longer than the length of the predefined time window, will be switched off during the duration. We conduct extensive trace-driven experiments to demonstrate its efficacy; on average, more than 70% of power consumption can be markedly saved with over 92% of WiFi coverage guaranteed, which is able to save empirical $59 000 per year just for our system.

With the significant advance of vehicle-to-everything (V2X) techniques, unsignalized intersection coordination has been widely recognized to facilitate the development of automated vehicles (AVs) for the intelligent transportation system. However, how to guarantee driving safety while improving the unsignalized intersection management efficiency is a challenging issue. In this article, we investigate the collision-free and efficient V2X-enabled AV scheduling problem at unsignalized intersections. First, by dividing the intersection zone into different collision sections (CSs), we formulate the intersection collision-free model into an absolute value programming (AVP) problem, which is proved to be NP-hard. We consider both nonplatoon and platoon traffic scenarios, and unlike previous algorithms, which require to control all the AVs at each scheduling step with computational intractability, our scheduling algorithm can assign a feasible time for each arriving AV with low complexity. Further, we propose an alternately iterative descent method (AIDM) to solve the AVP problem by assigning the optimal entering time for each arriving AV. Through extensive simulations with various traffic data generated by SUMO, we demonstrate that our proposed AIDM algorithm can significantly enhance the scheduling performance in terms of passing delay and scheduling throughput. Even though the AIDM algorithm achieves the same level of transportation performances with the state-of-the-art algorithm, it advances dramatically in computational complexity and communication overhead, which is easier to be implemented in practice.

Internet of Things (IoT) computing offloading is a challenging issue, especially in remote areas where common edge/cloud infrastructure is unavailable. In this paper, we present a space-air-ground integrated network (SAGIN) edge/cloud computing architecture for offloading the computation-intensive applications considering remote energy and computation constraints, where flying unmanned aerial vehicles (UAVs) provide near-user edge computing and satellites provide access to the cloud computing. First, for UAV edge servers, we propose a joint resource allocation and task scheduling approach to efficiently allocate the computing resources to virtual machines (VMs) and schedule the offloaded tasks. Second, we investigate the computing offloading problem in SAGIN and propose a learning-based approach to learn the optimal offloading policy from the dynamic SAGIN environments. Specifically, we formulate the offloading decision making as a Markov decision process where the system state considers the network dynamics. To cope with the system dynamics and complexity, we propose a deep reinforcement learning-based computing offloading approach to learn the optimal offloading policy on-the-fly, where we adopt the policy gradient method to handle the large action space and actor-critic method to accelerate the learning process. Simulation results show that the proposed edge VM allocation and task scheduling approach can achieve near-optimal performance with very low complexity and the proposed learning-based computing offloading algorithm not only converges fast but also achieves a lower total cost compared with other offloading approaches.

Full WiFi coverage is more and more prevalent in many places such as university, enterprise, big mall, etc. To achieve full WiFi coverage in a wide area is very costly. Not only extensive AP deployments are expensive, to operate and maintain such large-scale APs every day can also cost much, e.g., the huge power consumption. In this paper, we collect large-scale AP status data in our campus WiFi system, which contains over 8,000 APs and serves about 40,000 active end-users in the area of 3.0925 km2. After conducting empirical studies on AP loads, we find Idle Phenomenon prevails throughout the trace. A large portion of APs are running without any user association, which will inevitably lead to unnecessary energy consumption. Inspired by this, we propose an intelligent large-scale AP control scheme, named as ACE (i.e., AP Control with Energy saving), to dynamically control large-scale APs (On or Off for energy saving meanwhile without loss of WiFi coverage. In ACE, the load of each AP is predicted first by the random forest algorithm, and those APs whose idle durations last for more than the length of the pre-defined sliding window will be turned off. We conduct extensive trace-driven simulations to demonstrate the efficiency of the ACE scheme; specifically, more than 70% of power energy can be saved with over 92 % of user WiFi coverage guaranteed in average.

Classifying traffic flows into source applications is of great value for intelligent network management, which can help to detect malicious attacks, monitor the network, optimize network behaviors and then improve user experience, etc. However, to achieve high-accuracy traffic classification, especially in real time, is very challenging due to very complicated behaviors of traffic flows where network applications could often transmit traffics with encryption at randomized port numbers under highly dynamic network conditions. In this paper, by collecting extensive application traffic flows at the exit router of Shanghai Maritime University (the traffic rate can reach up to 7 GB/s at peak time), we identify that there is a very distinct characteristic in inner-connection of message (grouped by single or multiple consecutive TCP packets) sequence for different application flows. We then propose our traffic classification algorithm, which essentially adopts a Long Short-Term Memory (LSTM) neural network to output a classifier with message sequence vector (not necessarily covering all messages) of a traffic flow as the training input, to conduct online traffic flow classification. Extensive simulations are conduced considering varied training data size and diverse source applications, and an average about 97 % accuracy on per-flow classification can be achieved.

Vehicular ad hoc network (VANET) has been widely recognized as a promising solution to enhance driving safety, by keeping vehicles well aware of the nearby environment through frequent beacon message exchanging. Due to the dynamic of transportation traffic, especially for those scenarios where the density of vehicles is high, the naive beaconing scheme where vehicles send beacon messages at a fixed rate with a fixed transmission power can cause severe channel congestion. In this paper, we investigate the risk of rear-end collision model and define a danger coefficient ρ to characterize the danger threat of each vehicle being in a rear-end collision. We then propose a fully-distributed beacon congestion control scheme, referred to as ABC, which guarantees each vehicle to actively adapt a minimal but sufficient beacon rate to avoid a rear-end collision based on individual estimates of ρ. In essence, ABC adopts a TDMA-based MAC protocol and solves a NP-hard optimal distributed beacon rate adapting (DBRA) problem with a greedy heuristic algorithm, in which a vehicle with a higher ρ will be assigned with a higher beacon rate while keeping the total required beacon demand lower than the channel capacity. We conduct extensive simulations to demonstrate the efficiency of ABC design in different traffic density and a large variety of underlying road topologies.

Under the IEEE 802.11p-based dedicated short range communication modules, vehicular safety applications rely on periodical broadcasts of safety beacons by each vehicle. However, the channel can be easily congested by high-frequency periodic beacons when the vehicle density becomes heavy. In this paper, through real-trace-based empirical study on vehicle-to-vehicle communication, we find that nonline-of-sight (NLoS) condition is the key factor on link performance degradation and blindly sending more packets in harsh NLoS conditions can hardly succeed but increase interferences to neighboring vehicles. Inspired by this, we propose a distributed beacon congestion control (DBCC) scheme to control beacon activities with considering link conditions, i.e., vehicles with more neighbors and better conditions of links with its neighbors, will be assigned with higher beacon rates. In DBCC, we first utilize two machine learning methods, i.e., naive Bayes and support vector machines, to train the features and output a classifier model which conducts online NLoS link condition prediction. With link status information, we then formulate a link-weighted safety benefit maximization (L-SBM) problem of the rate-adaptation under a TDMA broadcast MAC, which is proved to be NP-hard. A greedy heuristic algorithm for L-SBM is then proposed and the performance of the algorithm is evaluated. Extensive trace-driven simulations demonstrate the efficiency of DBCC design; particularly, the rate of beacon transmissions can be effectively controlled without exceeding the resource limit and the rate of transmission/reception collisions are greatly reduced.

WiFi has been considered as the most promising radio technology to carry the rapidly growing in-vehicle Internet traffic, such as video streaming, user generated content sharing, etc. By offloading traffic from cellular networks to the roadside WiFi Access Points (APs), the traffic throughput can be improved with reduced network cost. Prior to Internet services, the vehicle has to accomplish the access procedure, which involves the transmission of the management frames such as probe request/response frames, authentication frames, etc. The access procedure can affect the throughput of the vehicular Internet connection, since the vehicle has to wait until all frames are successfully transmitted before accessing Internet services. To study the impact of such access procedure on the throughput performance of the in-vehicle Internet access via on-road WiFi APs, in this paper, we propose a two dimensional Markov chain model to study the progress of the access procedure when the vehicle drives through the consecutive zones within the AP coverage area. We evaluate the dependency of the throughput over different conditions, such as packet error rate, velocity of the vehicle, average packet delay, etc. The results of the paper will provide useful insights for future design and deployment of the roadside WiFi networks.