Energy Minimization for Intelligent Reflecting Surface-Assisted Mobile Edge Computing

Abstract

This paper presents a new mobile edge computing (MEC) framework, where an intelligent reflecting surface (IRS) is deployed to assist the smart terminals (STs) in offloading their computations to a base station (BS) equipped with edge servers. The energy consumption of the STs is minimized by jointly optimizing the CPU frequencies of the STs, the offloading schedule, the IRS phase shifts, and the receive beamformers of the BS. A key idea is that we reveal the optimal CPU frequency of each ST is time-invariant. Another important aspect is that given the optimal CPU frequency, we develop a new algorithm, which leverages alternating optimization to optimize the IRS phases and receive beamformers, and Lagrange duality method to optimize the offloading schedule. The proposed algorithm guarantees to compute a stationary point solution for the problem of interest with a low complexity. By using positive semidefinite relaxation (SDR) technique, we also develop a lower bound of the energy consumption. Numerical results show that the proposed algorithm reaches the lower bound, and is superior to the baseline schemes in terms of energy efficiency.