3 April 2023 in IEEE Internet of Things Journal
Multi-access edge computing (MEC) has emerged as a powerful paradigm for increasing the computation performance of mobile devices (MDs). Applying non-orthogonal multiple access (NOMA) to MEC can further improve the spectrum efficiency and reduce offloading delays caused by the upload congestion. In this paper, we examine the joint computation offloading and resource allocation problem in the NOMA-MEC system, which benefits from the combination of NOMA and MEC. Our optimization objective is to minimize the computational overhead (the weighted sum of the execution delay and the energy consumption) in dynamic environments with time-varying wireless fading channels. The optimization problem is formulated as a mixed integer programming (MIP), which involves jointly optimizing the task offloading decisions, channel assignment, and transmit power allocation. To solve such an optimization problem, we formalize the task offloading and the resource allocation as a Markov decision process (MDP). Then, we propose a deep reinforcement learning (DRL) based approach, which combines multiple deep neural networks (DNNs) to directly approximate different statistical models for continuous and discrete control. The simulation results demonstrate that the proposed approach can rapidly converge and efficiently decrease the total computational overhead compared to other baseline approaches in different scenarios.
30 August 2024 in IEEE Internet of Things Journal
Mobile edge computing (MEC) emerges as a promising paradigm that aims to extend the computing capabilities of mobile devices. This enhancement is achieved by offloading heavy computational tasks to nearby edge servers, where relevant services, such as programs, databases, and libraries, are stored to support the execution of these tasks. Given the restricted storage capacity and computing resources of edge servers, it is crucial to carefully consider which tasks to offload and which services to cache. However, the highly dynamic system and closely coupled decisions between offloading and caching pose considerable challenges to developing an effective strategy. In this article, we investigate both computation offloading and service caching in an end-edge–cloud collaborative system under time-varying wireless channels and task requests. We formulate an optimization problem aimed at minimizing the system cost, including time and energy consumption, and model it further as a Markov decision process (MDP). To address this, we propose a deep reinforcement learning (DRL)-based approach with dynamic action masking to concurrently optimize offloading and caching decisions. This approach renormalizes the valid action probability distribution at every step, enabling the agent to explore within the space of valid action combinations and thereby learn the optimal policy. Extensive experimental results demonstrate that the proposed approach greatly reduces the system cost compared to other representative benchmark schemes.