Resource Allocation in Multi-Access Edge Computing using Teaching-Learning Based Optimization: A Multi-Objective Approach

Abstract

Efficient resource allocation in Multi-access Edge Computing (MEC) plays a pivotal role in achieving high throughput, low latency, energy efficiency, and user fairness. Traditional optimization approaches often address these goals separately, leading to suboptimal solutions in dynamic environments with multifaceted user demands. This research proposes a multi-objective framework for resource allocation in MEC by leveraging the Teaching-Learning Based Optimization (TLBO) algorithm. The TLBO algorithm, inspired by the classroom learning process, iteratively improves a population of candidate solutions by sharing knowledge among learners and guidance from a 'teacher.' The research formulate the resource allocation problem as a multi-objective optimization problem and demonstrate how TLBO can effectively discover Pareto-optimal solutions that represent trade-offs between conflicting objectives. Experimental results on simulated MEC scenarios demonstrate the superiority of with throughput of 150 mbps the proposed approach compared to baseline strategies such as greedy of 135 mbps and weighted round robin of 142 mbps.