An Approach of Hopfield Network-Based Optimization for Task Offloading and Resource Allocation in Mobile Edge Computing

Abstract

Mobile edge computing (MEC) has become an effective solution to alleviate the delay and energy burdens of cloud computing infrastructures. By enabling task execution at edge servers located near user devices, MEC reduces transmission delay and improves responsiveness. Nevertheless, uncontrolled task offloading may degrade overall system performance, and optimizing task offloading and resource allocation (TORA) can ensure efficient workload distribution. Prior studies have primarily applied metaheuristic techniques such as genetic algorithms (GA) and particle swarm optimization (PSO), while machine learning-based solutions remain relatively limited. This work introduces a Hopfield network-based optimization model, where the TORA target is reformulated as an energy function, and a Hopfield network performs the minimization process. Simulation experiments confirm that the Hopfield approach reduces both delay and energy consumption more effectively than GA and PSO, with its longer runtime. However, this limitation can be mitigated by advances in modern processing power.