Search

Meeting the strict Quality of Service (QoS) requirements of terminals has imposed a significant challenge on Multi-access Edge Computing (MEC) systems, due to the limited multi-dimensional resources. To address this challenge, we propose a collaborative MEC framework that facilitates resource sharing between the edge servers, and with the aim to maximize the long-term QoS and reduce the cache switching costs through joint optimization of service caching, collaborative offloading, and computation and communication resource allocation. The dual timescale feature and temporal recurrence relationship between service caching and other resource allocation make solving the problem even more challenging. To solve it, we propose a deep reinforcement learning (DRL)-based dual timescale scheme, called DGL-DDPG, which is composed of a short-term genetic algorithm (GA) and a long short-term memory network-based deep deterministic policy gradient (LSTM-DDPG). In doing so, we reformulate the optimization problem as a Markov decision process (MDP) where the small-timescale resource allocation decisions generated by an improved GA are taken as the states and inputted into a centralized LSTM-DDPG agent to generate the service caching decision for the large-timescale. Simulation results demonstrate that the proposed algorithm outperforms the baseline algorithms in terms of the average QoS and the cache switching costs.

To provide satisfying services for ever-emerging mission-critical applications, the ultra-reliable and low-latency communications (URLLC) need novel design to improve the spectrum efficiency and enhance the robustness. To achieve this, we design a robust rate-splitting and beamforming scheme for the downlink multiuser URLLC system in finite blocklength regime under imperfect channel state information at the transmitter (CSIT) acquisition. Rate-splitting is utilized to deal with the complex inter-user interference and improve the spectrum efficiency. Considering the norm-bounded CSIT error model, we formulate a minimum user rate maximization problem to guarantee the URLLC performance requirements by jointly designing the rate-splitting factors and the common/private beamforming vectors. The corresponding constraints are infinite due to the uncertainty of CSIT and the constraint set is also non-convex. To tackle it, we convert the infinite constraints into finite ones utilizing S-Procedure, and transform the original problem into difference of convex (DC) programming. Efficient approaches based on constrained concave convex procedure and Gaussian randomization are proposed to solve the DC programming and generate initial feasible points. Through extensive simulations, the convergence, robustness and effectiveness proprieties of the design are investigated and confirmed. Compared with the baselines, our design can achieve obvious performance improvement for different blocklength and block error rate requirements.

Semantic communication has been identified as a core technology for the sixth generation (6G) of wireless networks. Recently, task-oriented semantic communications have been proposed for low-latency inference with limited bandwidth. Although transmitting only task-related information does protect a certain level of user privacy, adversaries could apply model inversion techniques to reconstruct the raw data or extract useful information, thereby infringing on users’ privacy. To mitigate privacy infringement, this paper proposes an information bottleneck and adversarial learning (IBAL) approach to protect users’ privacy against model inversion attacks. Specifically, we extract task-relevant features from the input based on the information bottleneck (IB) theory. To overcome the difficulty in calculating the mutual information in high-dimensional space, we derive a variational upper bound to estimate the true mutual information. To prevent data reconstruction from task-related features by adversaries, we leverage adversarial learning to train encoder to fool adversaries by maximizing reconstruction distortion. Furthermore, considering the impact of channel variations on privacy-utility trade-off and the difficulty in manually tuning the weights of each loss, we propose an adaptive weight adjustment method. Numerical results demonstrate that the proposed approaches can effectively protect privacy without significantly affecting task performance and achieve better privacy-utility trade-offs than baseline methods.

In unmanned aerial vehicle (UAV)-assisted multi-access edge computing (MEC) systems, where UAVs act as aerial relays to forward tasks from ground users (GUs) to remote edge servers (ESs) for processing, a crucial observation is that the computing power in the system depends on the computing capabilities at a single ES and the number of ESs covered by the UAV. The latter is essentially influenced by the UAV altitude, ES density, transmit power of the UAV, channel condition, etc. In this paper, we model a UAV-assisted MEC system featuring adjustable UAV altitude, random GU distribution, and random ES distribution. We adopt the signal-to-noise ratio-based coverage probability and derive a computing model to characterize communication-aware altitude-dependent computing power. Upon this, we model the sequential task-processing process, including task uploading, forwarding, and computing, as a three-stage tandem queue ( $M/D/1 \rightarrow D/1\rightarrow D/1$ ). Employing queueing theory, we derive analytical results for the end-to-end (e2e) service latency. Besides, we address the optimization problem of maximizing the number of completed tasks within the e2e latency constraint, referred to as task service throughput. Simulation and analytical results show that optimal UAV altitudes, yielding the maximum task computing throughput, can be obtained under given network parameters.

By encoding the video into different bitrate versions, dynamic adaptive streaming over HTTP (DASH) demonstrates its unique advantages in providing flexible bitrate adaption service in dynamic environments. But, the price is that the amount of video data is dramatically increased. The interaction of massive video data tends to exacerbate the network congestion and degrades the quality of experience (QoE) of users. Edge caching and mobile edge computing (MEC) have been adopted to solve this problem and enhance the QoE. But it is still difficult because of the highly coupled nature of caching and computing, which makes it extremely challenging to coordinate them across multiple edge nodes. To address the problem, this paper devotes itself to investigating collaborative edge caching and computing to maximize QoE for adaptive video streaming. In doing so, an optimization problem is formulated by jointly designing the caching, computing and user bitrate adaption, which turns out to be an integer nonlinear programming (INLP) problem and is NP-hard in strong sense. To solve it, we include caching placement, joint computing and bitrate adaption into a two-stage optimization framework. Specifically, considering the fact that the caching placement is implemented at a relatively long timescale, the caching problem is reformulated based on the statistics of user requests. The reformulated problem is a multiple-choice knapsack problem (MCKP), which is solved by Lagrange dual method after relaxation. The joint computing and bitrate adaption problem is transformed into Markov decision process (MDP) problem, and is solved by deep deterministic policy gradient (DDPG) algorithm. Simulation results validate that the proposed scheme can significantly improve QoE when compared with state-of-the-art baselines.

Reconfigurable intelligent surface (RIS) provides a promising way to proactively augment propagation environments for better transmission performance in wireless communications. Existing multi-RIS works mainly focus on link-level optimization with predetermined transmission paths, which cannot be directly extended to system-level management, since they neither consider the interference caused by undesired scattering of RISs, nor the performance balancing between different transmission paths. To address this, we study an innovative multi-hop multi-RIS communication system, where a base station (BS) transmits information to a set of distributed users over multi-RIS configuration space in a multi-hop manner. The signals for each user are subsequently reflected by the selected RISs via multi-reflection line-of-sight (LoS) links. To ensure that all users have fair access to the system to avoid excessive number of RISs serving one user, we aim to find the optimal beam reflecting path for each user, while judiciously determining the path scheduling strategies with the corresponding beamforming design to ensure the fairness. Due to the presence of interference caused by undesired scattering of RISs, it is highly challenging to solve the formulated multi-RIS multi-path beamforming optimization problem. To solve it, we first derive the optimal RISs’ phase shifts and the corresponding reflecting path selection for each user based on its practical deployment location. With the optimized multi-reflection paths, we obtain a feasible user grouping pattern for effective interference mitigation by constructing the maximum independent sets (MISs). Finally, we propose a joint heuristic algorithm to iteratively update the beamforming vectors and the group scheduling policies to maximize the minimum equivalent data rate of all users. Numerical results demonstrate that the proposed transmission framework achieves superior throughput performance than benchmark schemes. Useful insights on how to leverage multi-reflection paths over RISs to boost the throughput performance are also drawn under different settings for the multi-hop multi-RIS communication systems.

This paper investigates the dense multi-user sparse code multiple access (SCMA) cellular networks, where base stations (BSs) and users are randomly deployed in a 3D space by using stochastic geometry tools. First, a method increasing the dimension of the assignment matrix is presented to provide massive connectivity in each cell. The random pattern of arrangement of non-zero elements access (RPA) policy is utilized to allocate the available SCMA resources. Then the coverage probability is characterized under the nearest-BS (NBS) association mechanism by developing a multi-user connectivity model based on the RPA policy and average connectivity ratio (ACR). A compact closed-form expression of the coverage probability and its approximation are derived by utilizing the Toeplitz matrix and finite sum of power series, respectively. We then extract the area potential spectral efficiency (APSE) based on the coverage probability, and provide new insights into the 3D dense multi-cell SCMA networks. Simulation results confirm the precision of the theoretical results and demonstrate the advantage of the SCMA scheme.

With the development of computation-intensive applications, unmanned aerial vehicle (UAV)-enabled multi-access edge computing (MEC) provides task offloading service for the users with or without terrestrial infrastructure support. Meanwhile, the next generation of UAVs communication systems are expected to be user-centric. Therefore, more attention should be paid to users’ satisfaction with the offered service. In this paper, we study the service satisfaction-oriented task offloading and UAV scheduling problem for UAV-enabled MEC networks, where the task priorities are considered based on the delay requirements of users’ tasks and the remaining energy status of users. Specifically, we firstly divide the users into different groups via the K-means-based grouping algorithm. Then, we develop a novel user satisfaction model by jointly considering the task processing delay and energy saving, based on which a total user satisfaction maximization problem is formulated to jointly optimize the task offloading decisions and UAV scheduling strategy. To solve the formulated problem, we decompose it into two sub-problems, i.e., the UAV scheduling sub-problem and the task offloading sub-problem. To solve the first sub-problem, we develop a genetic algorithm (GA)-based UAV scheduling algorithm through dealing with multiple balanced assignment problems. To address the second sub-problem, a GA-based task offloading algorithm is developed. Then, we propose a joint task offloading and UAV scheduling optimization algorithm to solve the original optimization problem. Finally, simulation results demonstrate that the proposed optimization algorithm not only converges fast, but also improves the total users’ satisfaction greatly. The total users’ satisfaction improved by the proposed scheme is up to 16.9% in the case of 170 users.

This work investigates proactive edge caching for D2D-assisted wireless networks, where user equipments (UEs) can be selected as caching nodes to assist content delivery. The objective of this work is to achieve a trade-off between the cost for providing caching services and the content transmission latency. Doing so, there are two challenges: 1) Which UEs can be selected as caching nodes; 2) How to place contents on these selected UEs without user’s privacy disclosure. To address these, a novel community detection and attention-weighted federated learning based proactive edge caching (CAFLPC) strategy is proposed. In the strategy, we first group UEs into different communities based on both the mobility and social properties of UEs, and then select important users (IUs) as caching nodes for each community by considering the social importance of UEs. To determine how to place the popular contents in these selected IUs, an attention-weighted federated learning (AWFL) based content popularity prediction framework is proposed. It integrates the attention-weighted federated learning with Bidirectional Long Short Term Memory Network (AWFL_BiLSTM) to achieve a higher content popularity prediction accuracy while protecting user’s privacy. Considering the imbalance of UEs’ active levels and local computing capacities, an attention-weighted aggregation mechanism is proposed to improve the training efficiency and prediction accuracy. Simulations results show that the proposed CAFLPC strategy outperforms the compared existing caching strategies at about 2.2%-35.1% in terms of the transmission latency reduced by per unit cost.

Edge caching has emerged as a promising technique against latency caused by explosive growth of mobile data traffic through caching popular contents at the edge networks. However, the dynamically changing content popularity nature and limited caching capacity make it challenging to design an effective caching scheme to reduce latency. To solve this, a learning-based hierarchical edge caching (LHEC) scheme is proposed in this work. We first propose a novel deep learning architecture, namely Stacked Autoencoder-Long Short Term Memory Network (SAE-LSTMNet) to capture both the correlation of the request patterns among different content and the periodicity in time domain to improve the prediction accuracy of the content popularity. Then, to predict the popularity of these newly-added contents, a dynamic content catalog is introduced and a similarity-based content popularity prediction (SCPP) approach is proposed. Based on the content popularity prediction, a hierarchical edge caching optimization problem is formulated to minimize the average content downloading latency. Since the formulated problem is NP-hard and difficult to be solved, a low-complexity algorithm is proposed to obtain the near-optimal solutions. Simulation results show that the proposed content popularity prediction approach outperforms up to 6.36% in terms of the mean absolute error compared with the state-of-the-art methods and the proposed LHEC scheme reduces the average downloading latency at about 5.3%~7.9% compared with those existing caching schemes.