Multi-access Edge Computing Network Research Articles

SCMA-Based Multiaccess Edge Computing in IoT Systems: An Energy-Efficiency and Latency Tradeoff

Sparse code multiple access (SCMA) is a kind of code-domain non-orthogonal multiple access (NOMA) scheme, which can support the increasing requirements for high spectral efficiency and massive connections. Meanwhile, multi-access edge computing (MEC) is a promising technology for providing resource-constrained users with computing resources. In this paper, we propose a novel optimization scheme in SCMA-based MEC network from the perspective of energy and latency for the Internet of Things (IoT) devices. Specifically, a system utility is first used to calculate the weighted energy consumption and task execution latency. The initial utility minimization problem is non-convex and then can be subdivided into two tractable subproblems by fixing task offloading decisions, namely, optimal local computing via CPU frequency scheduling and optimal edge computing via SCMA codebook assignment, subcarrier power allocation, and MEC server computing resources distribution. Primarily, a joint SCMA codebook assignment based on the bidirectional matching principle and optimal power allocation algorithm is proposed. Moreover, we come up with CPU frequency scheduling strategies utilizing convex optimization to optimize the computing resources allocation of local devices and the MEC server. Finally, a low-complexity task offloading policy based on simulated annealing is presented. Numerical results show that our proposed joint optimization algorithm for resource allocation and task offloading can achieve a good compromise between time delay and energy consumption for IoT devices. It is demonstrated that the proposed strategy has a remarkable advantage compared to previous SCMA-MEC schemes.

HiveMind: Towards Cellular Native Machine Learning Model Splitting

The increasing processing load of today’s mobile machine learning (ML) application challenges the stringent computation budget of mobile user equipment (UE). With the wide deployment of 5G edge-cloud, a new ML offloading scheme called split ML is provisioned to enable computation-intensive mobile ML applications by splitting an ML model across mobile UE, edge, and cloud. However, the complex split assignment problems pose new challenges for split ML system design. In this paper, we introduce HiveMind, the first practical multi-split ML system tailored for 5G cellular networks. HiveMind reformulates the complicated multi-split problem to a min-cost graph search and optimizes the distributed algorithm to drastically reduce the signaling overhead. Benefit from its low overhead property, HiveMind makes the optimal split decision on multiple computing nodes in real-time and adapts the split decisions to the instantaneous network dynamics. HiveMind also incorporates a multi-objective mechanism that accommodates heterogeneous objectives for a single ML task. HiveMind adapts to a wide range of ML frameworks, including non-linear models like Recurrent Neural Network (RNN), Federated Learning (FL), and Multi-agent Reinforcement Learning (MARL). We evaluate HiveMind on 5G MEC network simulators with realistic traffic patterns and real-life MEC computation/communication profiles. Our experiments demonstrate that HiveMind achieves the optimal efficiency comparing to state-of-art split ML designs.

Low-latency service function chain migration in edge-core networks based on open Jackson networks

Multi-access Edge Computing (MEC) offers cloud computing capabilities at the edge of the network. Growing demand for low-latency services requires Service Function Chains (SFCs) to be scaled up beyond MEC network to core network. To adapt to network dynamics and provide low-latency services, being able to migrate SFCs when needed is of paramount importance. However, migration of SFCs among edge and core networks such that average latency is optimized as well as considering resource consumption is an intractable challenge because improper migration of Virtual Network Functions (VNFs) results in failure of meeting the requirements of network policies. In this paper, we investigate SFCs in edge-core networks and model the Latency-aware Edge-Core SFCs Migration problem based on open Jackson networks. Two SFC migration algorithms, i.e., Profit-driven Heuristic Search (PHS) and Average Utilization Based (AUB), are proposed to efficiently optimize average latency of all SFCs in edge-core networks. Extensive evaluation results show that PHS optimizes average latency by 19.5%, while AUB can further reduce average latency by up to 36.9% by allowing a marginally higher number of VNF migrations.

5G-Enabled MEC: A Distributed Traffic Steering for Seamless Service Migration of Internet of Vehicles

Multiaccess edge computing (MEC) is considered as a backbone for the 5G network. The successive MEC network combines the networking and computation at the edge of the network to achieve the Quality of Services (QoS) with ultralow latency. The devices with mobility feature, whether hand-held devices or vehicles move from one edge server (ES) location to another ES, creates a nonoptimal environment in the long run. To maintain QoS and avoid service disruptions, existing network topologies do not fulfill the requirement. Hence, a unique traffic steering with dynamic path selection is required for live service migration of time-sensitive applications. In this article, we are the first to introduce a distributed traffic steering through the differentiation of two different types of network elements (i.e., ESs and routers), in a large MEC system. Using this concept, we, for the first time, resolve the scalability problem of a large MEC network into a partitioned MEC network. The proposed framework bounds the path-finding procedure with a filter strategy based on the network distance to eliminate the excess of nonrelated network elements. With a decentralized framework for MEC, we propose matrix-based dynamic shortest path selection and matrix-based dynamic multipath searching algorithms for dynamic path selection under the proposed autonomous network boundary discovery and must connect node block benchmarks. Our proposed dynamic traffic steering system works under two most important metric measurements (time delay and available bandwidth). Experimental results validate the effectiveness of dynamic and adaptive path searching in a partitioned controlled MEC network that significantly outperforms the centralized approaches with 35%–70% efficiency in QoS.

Performance analysis of multi-access edge computing networks for heterogeneous IoT systems

The emergence of Multi-access Edge Computing (MEC) has drastically transformed the computing capabilities of communication systems. By facilitating the convenience of low latency and on-demand services for delay-sensitive tasks, these platforms have also contributed towards mitigating the communication overheads to a larger extent. An emerging application area which has largely benefitted with the adoption of MEC is the Internet of Things (IoT) paradigms. However, with the exponential rise in interconnected IoT devices requesting provisioning of resources to the MEC may lead to traffic bottlenecks in the MEC networks. This further leads to lower response times and degradation in the network’s Quality of Service (QoS). In view of modeling request processing mechanism and monitoring of network QoS, a Markov process driven framework to model MEC-based IoT systems is proposed. By exploiting the analytical inference from pure birth–death processes and queue theoretic models the steady-state solutions for MEC networks is derived. Further, critical performance metrics like mean number of requests in the network, traffic intensity of requests, fluctuations in number of requests, and the mean response time of the network are provided. The corresponding results obtained for the above analysis are validated through numerical illustrations to prove the convergence of the framework and its impact on different network parameters.

Machine and Deep Learning for Resource Allocation in Multi-Access Edge Computing: A Survey

With the rapid development of Internet-of-Things (IoT) devices and mobile communication technologies, Multi-access Edge Computing (MEC) has emerged as a promising paradigm to extend cloud computing and storage capabilities to the edge of cellular networks, near to IoT devices. MEC enables IoT devices with limited battery capacity and computation/storage capabilities to execute their computation-intensive and latency-sensitive applications at the edge of the networks. However, to efficiently execute these applications in MEC systems, each task must be properly offloaded and scheduled onto the MEC servers. Additionally, the MEC servers may intelligently balance and share their computing resources to satisfy the application QoS and QoE. Therefore, effective resource allocation (RA) mechanisms in MEC are vital for ensuring its foreseen advantages. Recently, Machine Learning (ML) and Deep Learning (DL) have emerged as key methods for many challenging aspects of MEC. Particularly, ML and DL play a crucial role in addressing the challenges of RA in MEC. This paper presents a comprehensive survey of ML/DL-based RA mechanisms in MEC. We first present tutorials that demonstrate the advantages of applying ML and DL in MEC. Then, we present enabling technologies for quickly running ML/DL training and inference in MEC. Afterward, we provide an in-depth survey of recent works that used ML/DL methods for RA in MEC from three aspects: (1) ML/DL-based methods for task offloading; (2) ML/DL-based methods for task scheduling; and (3) ML/DL-based methods for joint resource allocation. Finally, we discuss key challenges and future research directions of applying ML/DL for resource allocation in MEC networks.

Pareto Optimality in Infinite Horizon Mean-Field Stochastic Cooperative Linear-Quadratic Difference Games

This paper is concerned with the mean-field stochastic cooperative linear-quadratic (LQ) dynamic difference game in an infinite time horizon. First, the necessary and sufficient conditions for the stability in the mean-square sense, and the stochastic Popov-Belevith-Hautus (PBH) eigenvector tests for exact observability and exact detectability of mean-field stochastic linear difference systems are derived by the <inline-formula><tex-math notation="LaTeX">$\mathscr {H}$</tex-math></inline-formula>-representation technique. Second, the relation between the solvability of the cross-coupled generalized Lyapunov equations&#x00A0;(CC-GLEs) and exact observability, exact detectability, and stability of the mean-field dynamic system is well characterized. It is then shown that the cross-coupled algebraic Riccati equations&#x00A0;(CC-AREs) admit a unique positive definite (positive semi-definite, respectively) solution under exact observability (exact detectability, respectively), which is also a feedback stabilizing solution. Furthermore, all Pareto optimal strategies and solutions can be respectively derived via the solutions to the weighted CC-AREs (WCC-AREs) and the weighted cross-coupled algebraic Lyapunov equations&#x00A0;(WCC-ALEs). Finally, a practical application on the computation offloading in the multi-access edge computing network (MECN) is presented to illustrate the proposed theoretical results.

Nonlinear Traffic Data Reconstruction in Large-Scale Internet of Vehicle Systems: A Neural Network Approach

Multi-Access Edge Computing (MEC) can reduce transmission costs and provide faster interactive responses in cloud server-centric Internet of Vehicle (IoV) scenarios. However, the resource space of MEC server is limited, and the large amount of data uploaded by RSU may limit the further development of MEC network. Compressed Sensing (CS) proves that for the data with certain internal structure, a complete data set can be reconstructed even if the sampling frequency is lower than the requirements of the sampling theorem. In the current works, traffic data compression is accomplished by directly reducing the sampling frequency in the process of collecting traffic data. However, CS reconstruction only utilizes the linear characteristics of data. Aiming to fully utilize the nonlinear characteristics of data, we propose a neural network based traffic data reconstruction model framework, called NRTD. NRTD instantiates the process of generalized matrix factorization and combines the potential features of multilayer perceptrons to improve the fitting degree of the model to the traffic flow data change curve. Parameter optimization and model comparison experiments are carried out with different missing rates and missing scenes. The experimental results show that the proposed method is superior to the conventional compressed sensing reconstruction model before improvement under different conditions.

Task Offloading and Trajectory Scheduling for UAV-Enabled MEC Networks: An Optimal Transport Theory Perspective

In this letter, we propose a task offloading framework for UAV-enabled MEC networks, where all smart mobile devices are scattered randomly following an arbitrary distribution. By exploiting the hovering and mobilizing capability of a UAV, task offloading service can be provided to the devices in all sub-areas of the network. To extend UAV operating time and associated network lifetime, we formulate an optimization problem to minimize the total energy consumption of the UAV through joint region partitioning and UAV trajectory scheduling. To solve it, we decompose it into two independent sub-problems, i.e., region partitioning and UAV trajectory optimization. For the first sub-problem, we model it as a semi-discrete optimal transport problem by considering the traffic balance among different sub-areas and propose an iterative algorithm to achieve the optimal solution. Then, the UAV trajectory optimization sub-problem is modeled as a traveling salesman problem to determine the shortest route. Simulation results show that the proposed scheme can significantly reduce the energy consumption of the UAV while achieving proper load balance.

A Multi-Uav and AP Cooperating Mec Network for Minimizing the Weighted Energy Consumption with Flight Trajectories Optimized

A Multi-Uav and AP Cooperating Mec Network for Minimizing the Weighted Energy Consumption with Flight Trajectories Optimized

DPRL: Task Offloading Strategy Based on Differential Privacy and Reinforcement Learning in Edge Computing

Mobile edge computing has been widely used in various IoT devices due to its excellent computing power and good interaction speed. Task offloading is the core of mobile edge computing. However, most of the existing task offloading strategies only focus on improving the unilateral performance of MEC, such as security, delay, and overhead. Therefore, focus on the security, delay and overhead of MEC, we propose a task offloading strategy based on differential privacy and reinforcement learning. This strategy optimizes the overhead required for the task offloading process while protecting user privacy. Specifically, before task offloading, differential privacy is used to interfere with the user’s location information to avoid malicious edge servers from stealing user privacy. Then, on the basis of ensuring user privacy and security, combined with the resource environment of the MEC network, reinforcement learning is used to select appropriate edge servers for task offloading. Simulation results show that our scheme improves the performance of MEC in many aspects, especially in security and resource consumption. Compared with the typical privacy protection scheme, the security is improved by 7%, and the resource consumption is reduced by 9% compared with the typical task offloading strategy.

A Collaborative Cache Strategy in Satellite-Ground Integrated Network Based on Multiaccess Edge Computing

Multiaccess edge computing (MEC) provides users with a network environment and computing storage capacity at the edge of the network, ensuring a deterministic service with low delivery delay. This paper introduces a new satellite-ground integrated collaborative caching network architecture based on MEC and studies the caching strategy. On the ground side, the edge nodes (ENs) are deployed to the user side to form a hierarchical collaborative cache mode centered on the base station. On the satellite side, we utilize intelligent satellite ENs to precache and multicast the highly popular contents, reducing the initial content delivery delay. Under the constraints of the user demand and storage capacity, we study the deployment and cache scheme of ENs and establish the delivery delay minimization problem. To solve the problem, we propose a content update decision parameter for content cache update and transform the problem into improving the hit rate of ENs. Simulation results show that the proposed MEC network architecture and content caching scheme can increase the caching system hit rate to 64% and reduce the average delay by 32.96% at most.

Joint computation and power allocation for NOMA enabled MEC networks in the finite blocklength regime

This paper investigates the reliable computation offloading in non-orthogonal multiple access enabled mobile edge computing networks, where the short-packet technique is adopted to meet the stringent latency requirements of delay-sensitive computing services. To characterize the reliability of computation offloading with finite blocklength coding, a novel analytical framework is developed to approximate the average overall block error probability (AOBEP) by using a double-case linearization. With the derived approximate expression for AOBEP, the joint optimization of computation workload and transmit power is further investigated, in order to minimize the AOBEP under the computation/communication delay constraints. To tackle the non-convexity of the formulated problem, the formulated problem is first decomposed into the computation workload problem and the transmit power allocation problem. For the computing workload problem, a closed-form solution is derived. Then, by applying the derived workload allocation, the power allocation is transformed into the convex form through some approximations and solved by the successive convex approximation technique. Finally, simulation results are provided to demonstrate the reliability enhancement of computation offloading and reveal the relationship between the workload/power allocation and the communication/computation delay.

Cluster Design and Optimization of SWIPT-Based MEC Networks with UAV Assistance

In recent years, service isolation and service miniaturization have become very popular. The large services are dismantled into multiple low-cost and simple small services to improve the scalability and disaster tolerance of the entire services. A service network composed of unmanned aerial vehicles (UAVs) and MEC servers is proposed in this paper, which aims at decoupling multiple services of the SWIPT-MEC network. In this network, UAVs take charge of energy transmission and calculation task scheduling and MEC servers are focused on task calculation. To meet the resource requirements of the ground nodes (GNs) in the network, we designed a distributed iterative algorithm to solve the resource allocation decision problem of GNs and used the modified expert bat algorithm to complete the UAV’s trajectory planning in a two-dimensional space. The results show that the algorithm can formulate a more fair resource allocation strategy, and its performance is improved by 7% compared with the traditional bat algorithm. In addition, the algorithm in this paper can also adapt to changes in task length and has a certain degree of stability.

Intelligent ubiquitous computing for future UAV-enabled MEC network systems

Intelligent ubiquitous computing for future UAV-enabled MEC network systems

A Multi-Leader Multi-Follower Stackelberg Game for Coalition-Based UAV MEC Networks

In this letter, we investigate the joint optimization of computation offloading, channel allocation and position deployment problem for coalition-based unmanned aerial vehicle (UAV) mobile edge computing networks. To characterize the hierarchical offloading feature in coalition-based UAV network, we propose the hierarchical framework and formulate a discrete multi-leader multi-follower energy minimization Stackelberg game. In addition, we prove that the multi-leader subgame and multi-follower subgame are exact potential games with Nash equilibrium. Then the existence of Stackelberg equilibrium (SE) is guaranteed. The spatial adaptive play-best response based hierarchical iterative learning (SAP-BRHIL) algorithm is proposed to achieve the SE. The simulation results show that the proposed SAP-BRHIL algorithm can achieve the lowest network energy consumption compared with some existing approaches.

Towards 5G: Joint Optimization of Video Segment Caching, Transcoding and Resource Allocation for Adaptive Video Streaming in a Multi-Access Edge Computing Network

Caching and transcoding at multi-access edge computing (MEC) server and wireless resource allocation in eNodeB interact with each other and together determine the quality of experience (QoE) of dynamic adaptive streaming over HTTP (DASH) clients. However, the relationship among the three factors has not been explored, which has led to limited improvement in clients’ QoE. Therefore, we propose a joint optimization framework of video segment caching and transcoding in MEC servers and resource allocation to improve the QoE of DASH clients. Based on the established framework, we develop an MEC caching management mechanism that consists of the MEC caching partition, video segment deletion, and MEC caching space transfer. Then, a joint optimization algorithm that combines the video segment caching and transcoding in the MEC server and resource allocation is proposed. In the algorithm, the clients’ channel state and the playback status and cooperation among MEC servers are employed to estimate the client's priority, video segment representation switch and continuous playback time. Considering the above four factors, we develop a utility function model of clients’ QoE. Then, we formulate a mixed-integer nonlinear programming mathematical model to maximize the total utility of DASH clients, where the video segment caching and transcoding strategy and resource allocation strategy are jointly optimized. To solve this problem, we propose a low-complexity heuristic algorithm that decomposes the original problem into multiple subproblems. The simulation results show that our proposed algorithms efficiently improve client's throughput, received video quality and hit ratio of video segments while decreasing the playback rebuffering time, video segment representation switch and system backhaul traffic.

Blockchain-Enabled Intelligent Video Caching and Transcoding in Clustered MEC Networks

In recent years, the number of smart devices has exploded, leading to an unprecedented increase in demand for video live and video-on-demand (VoD) services. Also, the privacy of video providers and requesters and the security of requested video data are much more threatened. In order to solve these issues, in this paper, a blockchain-enabled CMEC video transmission model (Bl-CMEC) for intelligent video caching and transcoding will be proposed to ensure the transactions’ transparency, system security, user information privacy, and integrity of the video data, enhance the ability of severs in actively caching popular video content in the CMEC system, and realize transcoding function at network edge nodes. Furthermore, we chose a scheme based on deep reinforcement learning (DRL) to intelligently access the intracluster joint caching and transcoding decisions. Then, the joint video caching and transcoding decision smart contract is specially designed to automatically manage the transaction process of the joint caching and transcoding service, which records key information of joint caching and transcoding transactions and payment information on a continuous blockchain. The simulation results demonstrate that the proposed Bl-CMEC framework not only can provide users with better QoE performance for video streaming service but also can ensure the security, integrity, and consistency for the video providers, video requesters, and video data.

Throughput Maximization of Offloading Tasks in Multi-Access Edge Computing Networks for High-Speed Railways

With the development of high-speed railways (HSRs), people's lives have become more convenient. At the same time, people are demanding more data services on trains in terms of wireless communications and mobile computation. Recently, Multi-access Edge Computing (MEC), a new computing paradigm, has the great potential to improve the user experience of computing services by placing the cloud infrastructures closer to end-users so that users can offload their computation-intensive and delay-sensitive tasks to the edge servers. In this paper, we concentrate on finding proper data routing paths for each offloaded task and each process result, with the objective of maximizing the network throughput by serving as many tasks as possible while meeting the MEC network resource constraints and the user demand constraints within a finite time horizon of equal time slots. It is challenging because of the unspecific data volume of an offloaded task and handovers of task offloading procedures that may frequently occur. Thus, we first investigate the impact of handovers in uplinks and downlinks. Then we propose a graph-based model to portray the resource availability in the network. Furthermore, a heuristic algorithm is devised, which finds a set of single-source-single-destination delay-constrained shortest paths in a series of auxiliary graphs. Finally, we perform an empirical evaluation of the proposed algorithm by simulations. The simulation results show that the proposed algorithm could achieve a higher network throughput than the other two benchmark heuristics.

Mobility-aware dynamic offloading strategy for C-V2X under multi-access edge computing

Multi-access edge computing (MEC) technology is envisioned as a promising paradigm to achieve the user needs of low-latency applications. Complex computation tasks are offloaded from resource-constrained devices to resource-rich devices to reduce task completion delays. However, given the high-speed mobility of vehicles, the traditional MEC network architecture is insufficient for the Internet of Vehicle (IoV) computing scenarios. Besides, many resource-rich idle vehicles can be used as mobile servers to perform computing tasks, extending the choices for vehicle users. To better integrate MEC and IoV technologies, we introduce a Cellular-V2X-based MEC offloading scenario, enabling MEC servers and vehicle nodes in the 5G cellular network environment to perform offloading cooperatively. For a typical delay-sensitive task, taking the Augmented Reality (AR) application as an example, we divide the task into multiple subtasks with linear correlation. A mobility-aware dynamic offloading algorithm (MADO) is proposed to minimize the impact of vehicular mobility on offloading. The algorithm is used to find an optimal resource allocation, which reduces the overhead of the entire offloading process. It continuously updates the resource selection and the offloading strategy with the vehicular location changes. Simulation results corroborate that our proposed MADO algorithm can effectively reduce task completion time, improve task completion success rate and adapt to a dynamic environment.