An Improved SD-Jaya Algorithm for Multi-Objective Edge Computing Problem in IRS-Aided Charging Electric Vehicular Networks

Nowadays, the connection of electric vehicles (EVs) to the grid leads to significant load growth, which is a challenge for the safe and stable operation of the grid. In vehicle edge computing networks, as grid-charged EVs offloading tasks, offloading strategy optimization is a crucial method of reducing power purchase from the grid by EVs. Therefore, we first construct an intelligent reflecting surface (IRS)-aided task offloading networks for charging EVs, innovatively employing tethered UAVs (tUAVs) equipped with IRS to transmit the task. Communication efficiency can be improved by choosing single-IRS or double-IRS links. In addition, EV is equipped with the network in box (NIB) to provide flexible computing power. Subsequently, the multi-objective problem of minimizing transmission obstacle, power purchase cost, and transmission delay is formulated. However, finding the Pareto front for such a multi-objective optimization problem in this paper is challenging. An improved SD-Jaya algorithm is proposed to obtain the selection of policy operators through historical knowledge. The simulation results illustrate that the proposed algorithm can get better non-dominated solutions.