Yong Ma

As a novel promising computational paradigm, wireless powered mobile edge computing (WPMEC) has been proposed to offer real-time energy and computing services for Internet of Things (IoT) devices. However, time-varying limited resources such as communication quality and residual energy pose great challenges in devising suitable real-time task offloading and resource allocation strategies to meet users' requirements for low latency and energy consumption. Existing studies either employ raw data offloading methods with significant communication overhead, or utilize ordinary data compression methods that result in poor compression effects. To cope with the challenges, this paper considers introducing the state-of-the-art lossless data compression technology into WPMEC and study the online joint optimization problem of task offloading decision, charging time allocation, and compression proportion allocation with the goal of optimizing task completion time. To tackle this problem, we propose an Intelligent Compression Offloading and adaptive resource Allocation algorithm called ICOA. We first put forward a well-devised framework based on deep reinforcement learning to generate a offloading decision vector set in real-time. Then we standardize the resource allocation problem as a linear programming problem and solve it using the simplex method. The experimental results on a real dataset show that, compared with the benchmark algorithms, the proposed algorithm can effectively reduce the task accomplishing time and energy consumption, and achieve the best approximate ratio. Moreover, ICOA requires low runtime, and can satisfy the real-time and effectiveness requirements very well.

As massive data is generated by Internet of Things (IoT) devices, user-end devices are required to implement computation-intensive functionalities, including multi-sensory data processing and analysis, sophisticated system control schemes, and artificial intelligence. Mobile Edge Computing (MEC) is a significant technology that has the potential to extend the computation and storage capacities of user-end devices by the decentralization of required resources and contents near users and at the edge. A crucial challenge in this direction is the development of a smart mechanism to effectively cache contents upon Edge Servers (ESs) near users for high effectiveness and low latency of content delivery with the constraints on computational and storage capacities of ESs. This study employs a queuing model for analyzing total request delay and interprets the content caching problem as an adversarial semi-bandits problem. We propose an Online Self-feedback Adversarial Semi-bandits Learning (OSAL) algorithm that incorporates a dual-layer learning architecture for dynamically generating caching strategies and maximizes the long-term reward. Experimental results demonstrate that the proposed method significantly outperforms the state-of-the-art methods across various performance metrics in a real-world multi mobile-user content caching case.

Detecting urban anomalies helps city managers solve traffic problems in a timely manner, prevent safety hazards, and make more scientific and rational urban development plans. However, most existing urban anomaly detection methods focus on single time slots and overlook the impact of anomalous events on traffic flow patterns. To this end, we propose a novel urban traffic flow pattern anomaly detection method in this paper, referred to as UTFP-AD. Our approach constructs traffic flow over a period of time as a temporal probability distribution and uses traffic flow dynamics to detect anomalous patterns. Specifically, we construct temporal probability distributions to represent regional traffic flow patterns and calculate the traffic flow dynamics to reflect traffic flow changes. Independent and correlated feature vectors are then input into the proposed Multi-One-Class Support Vector Machine to identify the final anomalous regions. For validation, we conducted experiments using both synthetic and real-world data, comparing our method with six existing methods. The experimental results demonstrate that UTFP-AD significantly outperforms the existing methods in terms of Recall, F1Score and hit rate.

Mobile edge computing (MEC) is an evolving paradigm for rendering services through network-accessible resources deployed over Internet of Things (IoT) nodes at the edge. Nevertheless, an MEC environment usually employs thousands of physical machines connected via hundreds of switches/routers that communicate and coordinate to deliver computing service. In such complicated systems, faults caused by software, human errors, and hardware are often unavoidable. The edge of network presents a dynamic environment with great quantities of terminals, high mobility of mobile devices, heterogeneous applications, and intermittent traffic. In such an environment, MEC can suffer from unbalanced resource provisioning and interruptions of faults occurring at different levels, which further causes task faults and affects service quality. To address this challenge, this work proposes a novel fault-tolerant offloading method for handling faults by leveraging a reinforcement-learning-based service offloading decision model. The model synthesizes a Dueling Deep Q Network (DQN)-based algorithm for deciding user offloading behaviors and an adaptive checkpointing method for improving task execution reliability. For the purpose of model validation and comparison, extensive simulations are conducted. Numerical results clearly demonstrate that the proposed method is highly effective and outperforms existing methods.

The prerequisite for implementing incentive mechanisms and reliable participant selection schemes in federated learning is to obtain the contribution of each participant. Available evaluation methods for participant contributions require the server to possess a test dataset, often impractical. Additionally, the excessively high complexity of these works is unacceptable when training complex models in large-scale federated learning system. To address these issues, we propose a lightweight contribution evaluation method for federated learning participants, named LCEFL, based on model projection theory, which does not require the server to provide a test dataset. In addition, a model compression method is designed to be used in LCEFL to reduce the computational complexity. Furthermore, a trusted aggregation method based on LCEFL is proposed, where the weight of each participant's local model is determined by its trust level, which can be calculated using its contribution evaluation result. Experimental results show that LCEFL can achieve nearly the same accuracy as schemes based on Shapley Value, while significantly reducing computational overhead by more than 50%. Compared to available aggregation methods, the proposed trusted aggregation scheme is able to accelerate the convergence speed of the global model and improve its accuracy by 2% to 45%.

Incentive plays an important role in mobile crowdsensing (MCS), as it impels mobile users to participate in sensing tasks and provide high-quality sensing data. However, considering the privacy (including identity privacy, sensing data privacy, and reputation value privacy) and practicality (including reliability, quality awareness, and efficiency) issues in practice, it is a challenge to design such an effective incentive scheme for MCS applications. Existing studies either fail to provide adequate privacy-preserving capabilities or have low practicality. To address these issues, we propose a scheme called BRRV in MCS which relies on two rounds of range reliability assessment to guarantee the reliability of data while achieving privacy preservation. In addition, we also present a lightweight scheme called LRRV in MCS which relies on a single round of range reliability assessment to guarantee the reliability of data while achieving lightweight and privacy preservation. Moreover, to fairly stimulate participants, constrain participants' malicious behavior, and improve the probability of high-quality data, we design a quality-aware reputation-based reward and penalty strategy to achieve dual incentives (including money incentives and reputation incentives) for participants. Furthermore, comprehensive theoretical analysis and experimental evaluation demonstrate that our proposed schemes are significantly superior to the existing schemes in several aspects.