Computation Offloading Strategy for Improved Particle Swarm Optimization in Mobile Edge Computing

Abstract

Mobile Edge Computing (MEC) reduces latency and energy consumption by migrating computing resources to the edge of the network. Computing offloading is one of the means to reduce latency and energy consumption in MEC. Reasonable offloading decisions can effectively reduce system cost. Aiming at the increase in system delay and energy consumption caused by the deployment of the MEC server in the 5G communication scenario, a computing offloading strategy EIPSO based on an improved particle swarm optimization (PSO) algorithm is proposed. Establish a delay, energy consumption and multiobjective optimization model, and for delay-sensitive mobile applications, the model is transformed into a delay minimization problem under energy consumption constraints, and a penalty function is added to balance delay and energy consumption. Through the proposed calculation offloading decision, the calculation task is reasonably allocated to the corresponding MEC server. The simulation results show that compared with the ALL-Local algorithm, MECR algorithm and PSAO algorithm, the total system cost of this algorithm is the smallest, and the EIPSO strategy can reduce the delay in the MEC and balance the load of the MEC server.