DRL-Based Multidimensional Resource Management in SWIPT-NOMA-Enabled MEC
- 25 August 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (4) , 3252-3266
Abstract
Mobile edge computing (MEC) enables communication users with limited computation power to offload computation-intensive tasks to the edge server, thus dramatically enhancing the limited computing capabilities of the users. As the reality of scarce spectrum resources and the energy-constrained nature of communication users, this paper introduces non-orthogonal multiple access (NOMA) and simultaneous wireless information and power transfer (SWIPT) techniques to achieve more efficient task offloading in MEC. To minimize the number of computationally failed tasks while simultaneously satisfying different quality of service (QoS) requirements of users, a joint resource management problem of the spectrum, computation, and energy resources is formulated. Due to the non-convexity of the offloading optimization problem and the stochastic nature of the constructed MEC environment, a multiple agents deep deterministic policy gradient (MADDPG)-based resource management algorithm is proposed to manage each user’s multidimensional resources without collaborating. The simulation results show that compared to other benchmark schemes, the proposed algorithm can effectively improve both the communication and computational performances in MEC.Keywords
Funding Information
- Key Project on Anhui Provincial Natural Science Study through the Colleges and Universities (KJ2020A0514)
- National Natural Science Foundation of China (62371087, 62371408, 62301467)
- Natural Science Foundation of Anhui Province (2308085QF198)
- Fundamental Research Funds for the Central Universities (20720220080)
This publication has 42 references indexed in Scilit:
- Applications of Deep Reinforcement Learning in Communications and Networking: A SurveyIEEE Communications Surveys & Tutorials, 2019
- Modeling and Analysis of Error Process in 5G Wireless Communication Using Two-State Markov ChainIEEE Access, 2019
- Simultaneous Wireless Information and Power Transfer (SWIPT): Recent Advances and Future ChallengesIEEE Communications Surveys & Tutorials, 2017
- Tasks scheduling and resource allocation in heterogeneous cloud for delay-bounded mobile edge computing2017
- Mobile Edge Computing: A Survey on Architecture and Computation OffloadingIEEE Communications Surveys & Tutorials, 2017
- Delay-optimal computation task scheduling for mobile-edge computing systems2016
- Internet of Things Session Management Over LTE—Balancing Signal Load, Power, and DelayIEEE Internet of Things Journal, 2015
- Human-level control through deep reinforcement learningNature, 2015
- To offload or not to offload? The bandwidth and energy costs of mobile cloud computing2013
- Energy-optimal mobile application execution: Taming resource-poor mobile devices with cloud clones2012