Mobile Edge Computing Resources Research Articles

In-Depth Analysis of MEC Resource Optimization and Reliability Under 5G Empowerment

<p>The swift proliferation of 5G networks underscores the pivotal role of Mobile Edge Computing (MEC) in catering to the ever-evolving service demands. This study introduces a dynamic resource orchestration framework tailored for MEC within 5G standalone cellular architectures. This framework addresses the intricate challenges of computational resource provisioning and power management in a holistic manner. Leveraging convex optimization principles, we devise an optimal resource allocation strategy that incorporates reliability metrics to ensure robust performance. Furthermore, we employ Deep Reinforcement Learning (DRL) techniques to tackle a formulated Markov Decision Process (MDP), aimed at optimizing resource distribution for latency minimization. Our simulation outcomes demonstrate the efficacy of the proposed approach in mitigating task delays and enhancing the system’s adaptability across diverse workload profiles and network environments, thereby contributing novel insights into the field with minimal overlap in existing literature.</p> <p> </p>

Optimal service caching, pricing and task partitioning in mobile edge computing federation

Mobile Edge Computing (MEC) federations aim to establish a joint edge service model between Edge Infrastructure Providers (EIPs) and clouds, facilitating the sharing and utilization of MEC services and resources. However, in such a hierarchical multi-EIP MEC federation environment, optimizing service caching, task partitioning, and pricing strategies to maximize the profits gained by EIPs and minimize the cost of mobile devices (MDs) is challenging. To address this challenge, This paper first formulates a two-stage multi-leader, multi-follower Stackelberg game between EIPs and MDs. Then, a new method combining a Stackelberg-based Multi-Agent Deep Deterministic Policy Gradient (STMADDPG) algorithm and a Transformer-based Popularity Prediction (TPP) model is devised to learn the optimal strategies. Extensive experiments with real-world datasets are conducted to demonstrate the excellent performance of the proposed STMADDPG-TPP method.

An Efficient Algorithm for Resource Allocation in Mobile Edge Computing Based on Convex Optimization and Karush–Kuhn–Tucker Method

Mobile edge computing (MEC) is receiving more attention than centralized cloud computing due to the massive increase in transmission and compute requirements in 5G vehicle networks. It offers a significant amount of processing and storage resources to the edge of networks, offloading applications from vehicle terminals that are computation-intensive and delay-sensitive. For devices with limited resources, it uses edge resources to provide computationally heavy operations while conserving energy. This paper proposes a novel approach for computing offloading in MEC. To effectively optimize the MEC resources, this paper proposes a novel algorithm. First, the joint optimization and service cache decision subproblems were determined from continuous and discrete variables. Then, the near-optimal solution is determined from the subproblems through convex optimization and Karush–Kuhn–Tucker method. Simulation results show that the proposed algorithm has better computational offloading and resource allocation performance as compared to existing algorithms.

Optimal Resource Provisioning and Task Offloading for Network-Aware and Federated Edge Computing.

Compared to cloud computing, mobile edge computing (MEC) is a promising solution for delay-sensitive applications due to its proximity to end users. Because of its ability to offload resource-intensive tasks to nearby edge servers, MEC allows a diverse range of compute- and storage-intensive applications to operate on resource-constrained devices. The optimal utilization of MEC can lead to enhanced responsiveness and quality of service, but it requires careful design from the perspective of user-base station association, virtualized resource provisioning, and task distribution. Also, considering the limited exploration of the federation concept in the existing literature, its impacts on the allocation and management of resources still remain not widely recognized. In this paper, we study the network and MEC resource scheduling problem, where some edge servers are federated, limiting resource expansion within the same federations. The integration of network and MEC is crucial, emphasizing the necessity of a joint approach. In this work, we present NAFEOS, a proposed solution formulated as a two-stage algorithm that can effectively integrate association optimization with vertical and horizontal scaling. The Stage-1 problem optimizes the user-base station association and federation assignment so that the edge servers can be utilized in a balanced manner. The following Stage-2 dynamically schedules both vertical and horizontal scaling so that the fluctuating task-offloading demands from users are fulfilled. The extensive evaluations and comparison results show that the proposed approach can effectively achieve optimal resource utilization.

Resource Management in Mobile Edge Computing: A Comprehensive Survey

With the evolution of 5G and Internet of Things technologies, Mobile Edge Computing (MEC) has emerged as a major computing paradigm. Compared to cloud computing, MEC integrates network control, computing, and storage to customizable, fast, reliable, and secure distributed services that are closer to the user and data site. Although a popular research topic, MEC resource management comes in many forms due to its emerging nature and there exists little consensus in the community. In this survey, we present a comprehensive review of existing research problems and relevant solutions within MEC resource management. We first describe the major problems in MEC resource allocation when the user applications have diverse performance requirements. We discuss the unique challenges caused by the dynamic nature of the environments and use cases where MEC is adopted. We also explore and categorize existing solutions that address such challenges. We particularly explore traditional optimization-based methods and deep learning-based approaches. In addition, we take a deeper dive into the most popular applications and use cases that adopt MEC paradigm and how MEC provides customized solutions for each use cases, in particular, video analytics applications. Finally, we outline the open research challenges and future directions. 1

Cost-Effective Task Offloading in NOMA-Enabled Vehicular Mobile Edge Computing

Nonorthogonal multiple access (NOMA) and mobile edge computing (MEC) are two key emerging technologies for vehicular networks, where NOMA allows multiple vehicular user equipments (VUEs) to share the same wireless resources, and thus to enhance the spectrum utilization and system capacity, and MEC permits VUEs to offload their complex applications to MEC servers, and thus to provide support for computationally intensive intelligent applications. In this article, a NOMA-based vehicle edge computing (VEC) network model is proposed, and the cost minimization problem is constructed. Under the premise of ensuring the delay tolerance of all VUEs, the total system cost is minimized through the joint optimization of offloading decision-making, VUE clustering, subchannel and computation resource allocation, and transmission power control. Since the proposed problem is a mixed-integer nonlinear programming problem, which is difficult to solve, we decouple it into two subproblems and propose two heuristic algorithms to solve the task offloading and the MEC resource assignment problem, respectively, and finally, we obtain the closed-form solutions for cloud-related optimization problems through simple analysis. Simulation results show that the proposed joint algorithm is superior to other baseline algorithms in terms of system cost minimization.

Energy-Efficient UAV-Mounted RIS Assisted Mobile Edge Computing

Unmanned aerial vehicle (UAV) and reconfigurable intelligent surface (RIS) have been recently applied in the field of mobile edge computing (MEC) to improve the data exchange environment by proactively changing the wireless channels through maneuverable location deployment and intelligent signals reflection, respectively. Nevertheless, they may suffer from inherent limitations, e.g., UAV’s low endurance and RIS’s finite coverage, which limit their wider applications. UAV-mounted RIS (U-RIS), as a promising integrated approach, can combine the advantages of UAV and RIS to break the limit. Inspired by this, we consider a novel U-RIS assisted MEC system, where a U-RIS is deployed to assist the communication between the ground users and an MEC server. The joint UAV trajectory, RIS passive beamforming and MEC resource allocation design is developed to maximize the energy efficiency (EE) of the system. To tackle the intractable non-convex problem, we divide it into two subproblems and solve them iteratively based on successive convex approximation (SCA) and the Dinkelbach method. Finally we obtain a high-performance suboptimal solution. Simulation results show that the proposed algorithm significantly improves the energy efficiency of the MEC system.

Risk-Averse Investment Strategy for MEC Service Provisioning: A Data-Driven Distributionally Robust Solution

The emerging Internet of Things (IoT) era has stimulated many new computation-intensive applications. To support them, mobile edge computing (MEC) is a promising solution that allows users to offload their heavy computing tasks to nearby edge servers. Taking such computation offloading as the service, application service providers (ASPs) can rent resources from mobile network operators for MEC service provisioning. However, it is challenging for ASPs to determine how many resources to rent at different regions and times due to the uncertain user demand. When making an investment strategy, it is crucial to maximize the profit with the consideration on the Quality of Service (QoS), where a joint scheduling on both communication and computing resource under the uncertain demand is needed. To deal with the uncertainty, the probability distribution information is usually employed, which, unfortunately, might be hardly obtainable in practice. Therefore, in this article, we propose a data-driven risk-averse MEC resource investment (DRAI) strategy, where the demand uncertainty issue is particularly addressed. Specifically, we formulate the DRAI strategy into a stochastic optimization problem, which can achieve the expected optimal profit under the QoS guarantee statistically from a risk-averse perspective. To solve it, instead of relying on specific distribution models, we construct an ambiguity set based on the statistical characteristics derived from the historical data that contains all possible distributions, and develop a data-driven distributionally robust solution, aiming at achieving the best strategy under the worst case to make it trustworthy. Simulation results illustrate the effectiveness of the proposed DRAI strategy.

An efficient computation offloading and resource allocation algorithm in RIS empowered MEC

Mobile edge computing (MEC) enables mobile devices (MDs) to offload computation-intensive tasks to edge servers to support a variety of latency-sensitive emerging applications (such as the Internet of Vehicles, real-time video analytics, etc.). However, the time-varying communication link environment of signal occlusion and interference between MDs and edge servers often leads to disappointing offloading benefits. Reconfigurable intelligent surface (RIS) is recognized as a promising technology in sixth-generation communication networks, with great potential to intelligently adjust the phase shift and amplitude of reflective elements to enhance wireless network capabilities. This paper proposes a novel computation offloading algorithm for RIS empowered MEC networks. Specifically, we comprehensively consider the optimization problems of delay, energy consumption, and operator cost in the process of computation offloading, and model it as a Markov decision process. To overcome the continuous action space challenge, we propose a computation offloading algorithm based on Deep Deterministic Policy Gradient (DDPG) to jointly optimize the phase shift and amplitude of RIS, offloading decision, and MEC resource allocation strategy. Finally, compared with various other benchmark algorithms, our proposed algorithm has a significant performance improvement over non-RIS learning algorithms and other classical algorithms, and maintains the optimal performance.

ECMS: An Edge Intelligent Energy Efficient Model in Mobile Edge Computing

With the increasing popularity of mobile edge computing (MEC) for processing intensive and delay sensitive IoT applications, the problem of high energy consumption of MEC has become a significant concern. Energy consumption prediction and monitoring of edge servers are crucial for reducing MEC’s carbon footprint in accordance with green computing and sustainable development. However, predicting energy consumption of edge servers is a nontrivial problem due to the fluctuation and variation of different loads. To address this problem, we propose ECMS, a new edge intelligent energy modeling approach that jointly adopts Elman Neural Network (ENN) and feature selection to optimize the consumption of energy on edge servers. ECMS considers 29 parameters relevant to edge server energy consumption and uses the ENN to develop an energy consumption model. Unlike other energy consumption models, ECMS can successfully deal with load fluctuation and various sorts of tasks, such as CPU-intensive, online transaction-intensive, and I/O-intensive. We have validated ECMS through extensive experiments and compared its performance in terms of accuracy and training time to several baseline approaches. The experimental results show the superiority of ECMS to the baseline models. We believe that the proposed model can be used by the MEC resource providers to forecast and optimize energy use.

A Survey on Mobile Edge Computing Infrastructure: Design, Resource Management, and Optimization Approaches

Emerging 5G cellular networks are expected to face a dramatic increase in the volume of mobile traffic and IoT user requests due to the massive growth in mobile devices and the emergence of new compute-intensive applications. Running high-intensive compute applications on resource-constrained mobile devices has recently become a major concern, given the constraints of finite computation and limited storage capacities. Mobile Edge Computing (MEC) has recently become the key technology to overcome these issues by providing cloud computing capabilities and placing IT infrastructures at the mobile network edge. In this survey, we present a list of relevant research papers for the MEC infrastructure implementation phases, including (1) MEC infrastructure designing and dimensioning, (2) MEC infrastructure virtualization using Network Function Virtualization (NFV) concept, and the use of virtualized service placement and auto-scaling methods to deploy an agile system framework, (3) MEC resource management frameworks, and (4) approaches used to optimize the MEC resources on the physical infrastructure. The main focus of this survey is to determine the required aspects to implement an auto-scaled and proactive MEC-NFV infrastructure to support a dynamic and heterogenous mobile users’ demand at mobile network operators.

Optimized Joint Allocation of Radio, Optical, and MEC Resources for the 5G and Beyond Fronthaul

In 5G and beyond telecommunication infrastructures a crucial challenge in achieving the strict Key Performance Indicators (KPIs) regarding capacity, latency, and guaranteed quality of service, is the efficient handling of the fronthaul bottleneck. This part of the next generation networks is expected to comprise the New Radio (NR) access and the Next Generation Passive Optical Network (NGPON) domains. Latest developments load the fronthaul with computing tasks as well (e.g., for AI-based processes) in the context of Mobile Edge Computing (MEC). Towards efficient management of all resource types, this paper proposes a joint allocation scheme with three optimization phases for radio, optical, and MEC resources. This scheme, which has been developed in the context of the blueSPACE 5G Infrastructure Public Private Partnership (5G PPP) project, exploits cutting-edge technologies, such as radio beamforming, spatial-spectral granularity in optical networks, and Network Function Virtualization (NFV), to provide dynamic, adaptive, and energy efficient allocation of resources. The devised model is mathematically described and the overall solution is evaluated in a realistic simulation scenario, demonstrating its effectiveness.

Spatio-temporal Bayesian Learning for Mobile Edge Computing Resource Planning in Smart Cities

A smart city improves operational efficiency and comfort of living by harnessing techniques such as the Internet of Things (IoT) to collect and process data for decision-making. To better support smart cities, data collected by IoT should be stored and processed appropriately. However, IoT devices are often task-specialized and resource-constrained, and thus, they heavily rely on online resources in terms of computing and storage to accomplish various tasks. Moreover, these cloud-based solutions often centralize the resources and are far away from the end IoTs and cannot respond to users in time due to network congestion when massive numbers of tasks offload through the core network. Therefore, by decentralizing resources spatially close to IoT devices, mobile edge computing (MEC) can reduce latency and improve service quality for a smart city, where service requests can be fulfilled in proximity. As the service demands exhibit spatial-temporal features, deploying MEC servers at optimal locations and allocating MEC resources play an essential role in efficiently meeting service requirements in a smart city. In this regard, it is essential to learn the distribution of resource demands in time and space. In this work, we first propose a spatio-temporal Bayesian hierarchical learning approach to learn and predict the distribution of MEC resource demand over space and time to facilitate MEC deployment and resource management. Second, the proposed model is trained and tested on real-world data, and the results demonstrate that the proposed method can achieve very high accuracy. Third, we demonstrate an application of the proposed method by simulating task offloading. Finally, the simulated results show that resources allocated based upon our models’ predictions are exploited more efficiently than the resources are equally divided into all servers in unobserved areas.

Mobile edge computing resource allocation: An operating system view

In this paper, we study the mobile edge computing (MEC) resource allocation problem from the view of operating systems (OSs). To be more specific, we consider a multi-OS MEC resource allocation problem, where end users (EUs)/fog nodes (FNs) have different OSs. The main obstacle of the problem is the heterogeneity. First, for EUs, according to their OSs’ properties, EUs can be mainly categorized into soft real-time EUs and hard real-time EUs. Second, there exists different OSs for FNs. Thus, a unified FNs’ qualification constraint may be in need to develop. This constraint should be applicable for diverse FNs. Third, although the conventional proportional fairness is a universal fairness criterion for the service provider (SP) to allocate resources for different EUs, the specified proportions which guarantee the service quality are seemed hard to be developed. Therefore, a user satisfaction mechanism (USM) is proposed to first denote the different requirements between soft real-time and hard real-time applications of EUs, and second test whether FNs are qualified for the MEC service. In addition, a novel p-fairness is designed for the differentiated and quality-guaranteed services provided by SPs. Then, the USM and p-fairness driven MEC resource allocation problem is formulated as a mixed integer non-linear programming (MINLP) problem, where we propose a variable-substitution convexity transformation (VSCT) scheme to tackle it. In VSCT, we first utilize the variable substitution to transform the integer variables’ quadric constraints to linear constraints. Then, we relax several constraints and the problem is transformed as a convex optimization problem which is solved iteratively. Finally, we analyze the upper and lower bounds of the proposed VSCT. In the simulation, we discuss the performances of USM and p-fairness (i.e., user satisfaction). Moreover, we compare VSCT with the existing optimization method (i.e., branch & bound). Finally, we demonstrate some interesting properties of our proposed scheme.

Deep reinforcement learning for dynamic computation offloading and resource allocation in cache-assisted mobile edge computing systems

Mobile Edge Computing (MEC) is one of the most promising techniques for next-generation wireless communication systems. In this paper, we study the problem of dynamic caching, computation offloading, and resource allocation in cache-assisted multi-user MEC systems with stochastic task arrivals. There are multiple computationally intensive tasks in the system, and each Mobile User (MU) needs to execute a task either locally or remotely in one or more MEC servers by offloading the task data. Popular tasks can be cached in MEC servers to avoid duplicates in offloading. The cached contents can be either obtained through user offloading, fetched from a remote cloud, or fetched from another MEC server. The objective is to minimize the long-term average of a cost function, which is defined as a weighted sum of energy consumption, delay, and cache contents' fetching costs. The weighting coefficients associated with the different metrics in the objective function can be adjusted to balance the tradeoff among them. The optimum design is performed with respect to four decision parameters: whether to cache a given task, whether to offload a given uncached task, how much transmission power should be used during offloading, and how much MEC resources to be allocated for executing a task. We propose to solve the problems by developing a dynamic scheduling policy based on Deep Reinforcement Learning (DRL) with the Deep Deterministic Policy Gradient (DDPG) method. A new decentralized DDPG algorithm is developed to obtain the optimum designs for multi-cell MEC systems by leveraging on the cooperations among neighboring MEC servers. Simulation results demonstrate that the proposed algorithm outperforms other existing strategies, such as Deep Q-Network (DQN).

Resource Provision and Allocation Based on Microeconomic Theory in Mobile Edge Computing

Mobile edge computing (MEC) can significantly improve the performance of mobile applications by leveraging nearby servers as the edge cloud to provide task offloading execution service for a smart mobile device (SMD) through wireless access points (APs). However, the edge cloud and AP will not provide free services. Their radio frequency resources and computing resource are limited but the service requests from various mobile devices could be massive. The goal of this article is to provide a pricing mechanism to efficiently allocate limited resources in the MEC system according to the budget of SMDs. To this end, we first present a market model of MEC resources that can give a real insight into the incentives for resource sharing at network edges. In the model, computation, and radio resources can be traded between resource suppliers (AP and edge cloud) and buyers (SMDs). Furthermore, we employ the microeconomic theory to get an optimal budget allocation strategy for the SMD to maximize its utility within a limited budget. Moreover, we propose an Equilibrium Price Finding (EPF) algorithm to find the equilibrium price of the MEC system, maximizing the whole system utility and leading to optimal resource allocation. Finally, simulation results show that, compared with state-of-the-art resource allocation methods, our optimal budget allocation algorithm can find budget allocation strategy more effectively and our equilibrium price finding algorithm can achieve market equilibrium to optimally allocate computation and radio resources in the MEC system.

Decomposable Intelligence on Cloud-Edge IoT Framework for Live Video Analytics

With the rapid development of deep learning technology, the modern Internet-of-Things (IoT) cameras have very high demands on communication, computing, and memory resources so as to achieve low latency and high accuracy live video analytics. Thanks to the mobile-edge computing (MEC), intelligent offloading to the MEC nodes can bring a lot of benefits, especially when the decomposable pipeline is adopted in the cloud-edge architecture. In this article, we provide decomposable intelligence on a cloud-edge IoT (DICE-IoT) framework to support joint latency- and accuracy-aware live video analytic services. Specifically, the intelligent framework enables the pipeline-sharing mechanism to reduce MEC resource usage. A Nash bargaining is proposed to incentivize cooperative computing provision between the MEC and the cloud, and a generalized benders decomposition (GBD)-based approach is utilized to optimize the social welfare. The results show that the proposed DICE-IoT framework can achieve a win–win–win solution to the IoT device, the MEC, and the cloud stratum.

CE-IoT: Cost-Effective Cloud-Edge Resource Provisioning for Heterogeneous IoT Applications

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

A Learning Based Framework for MEC Server Planning With Uncertain BSs Demands

Mobile Edge Computing (MEC) architecture is composed of geographically distributed edge servers, in which computing capabilities are provisioned at the boundary of the network, which is in close proximity to the end users to provide network services with low latency. The planning of MEC edge servers at appropriate locations is the fundamental first step towards the deployment of the MEC system. In the literature, edge servers planning is based on deterministic resource requirements. This assumption largely neglects the pragmatic complexities imposed by the real dynamic world, in which base station (BS) resource demands are stochastic variables with arbitrary pattern. In view of this fact, we formulate the MEC planning problem as a joint optimization problem of MEC edge servers placement and resource allocation with uncertain BS demands through an uncertain programming formulation. Due to the complexity of this joint-uncertain problem, a learning based framework is utilized to practically solve this problem, and the relevance of applying this mechanism in practical usage with sampled arbitrary BS demands data is also discussed. Finally, we conducted intensive real-data driven simulations to evaluate the performance of our proposed mechanism. The results show the effectiveness of our approach with arbitrary BS demands.

Joint Offloading and Charge Cost Minimization in Mobile Edge Computing

Mobile edge computing (MEC) brings a breakthrough for Internet of Things (IoT) for its ability of offloading tasks from user equipments (UEs) to nearby servers which have rich computation resource. 5G network brings a huge breakthrough on transmission rate. Together with MEC and 5G, both execution delay of tasks and time delay from downloading would be shorter and the quality of experience (QoE) of UEs can be improved. Considering practical conditions, the computation resource of an MEC server is finite to some extent. Therefore, how to prevent the abuse of MEC resource and further allocate the resource reasonably becomes a key point for an MEC system. In this paper, an MEC system with multi-user is considered where a base station (BS) with an MEC server, which can not only provide computation offloading service but also data cache service. Especially, we take the charge for both data transmission and task computation as one part of total cost of UEs, and then explore a joint optimization for downlink resource allocation, offloading decision and computation resource allocation to minimize the total cost in terms of the time delay and the charge to UEs. The proposed problem is formulated as a mixed integer programming (MIP) one which is NP-hard. Therefore, we decouple the original problem into two subproblems which are downlink resource allocation problem and joint offloading decision and computation resource allocation problem. Then we address these two subproblems by using convex and nonconvex optimization techniques, respectively. An iterative algorithm is proposed to obtain a suboptimal solution in polynomial time. Simulation results show that our proposed algorithm performs better than benchmark algorithms.