Edge Computing Offload Optimization Strategy Based on Improved Particle Swarm

In multi-access edge computing, it is crucial to use computing offloading to ensure low processing latency for the task of the user device. This implies that efficient computing offloading is an effective way to reduce the average task latency problem. Therefore, this paper proposes an offloading strategy for multi-user multi-edge server scenarios by using an improved particle swarm. First, it takes transmission latency, processing latency, and the number of users into account when designing an objective function to minimize the average task latency. Then, the improved particle swarm algorithm with dynamic weights and learning factors is used to optimize the task offloading strategy. The simulation results show that the proposed algorithm can find a better offloading strategy compared with the classical particle swarm algorithm and the genetic algorithm.