Joint Optimization of Transmission and Computing Resource in Intelligent Reflecting Surface-Assisted Mobile-Edge Computing System

Abstract

Intelligent Reflecting Surface (IRS) is a promising approach to effectively improve the propagation environment, which includes a controller and numerous reflecting elements. In this paper, we consider an IRS-assisted mobile edge computing (MEC) system, which has a base station (BS), multiple single-antenna user terminals (UTs), and an IRS. Aiming at minimizing the system energy consumption, the transmission power of UTs, the BS receiving beamforming vector, the BS computing resources allocation, and the IRS effective phase shifts are jointly optimized. As these four variables are coupled together and the problem is non-convex, block coordinate descent method is adopted to decompose the optimization problem into four subproblems. In order to address the transmission power subproblem, quadratic transform based fractional programming, Lagrange dual transformation, and difference of convex function algorithm are used. Quadratic transformation and Lagrange dual transformation are also used to optimize the phase shift matrix and the receiving beamforming vector, while the quadratic transform in the multidimensional and complex case is used additionally in the IRS phase-shift subproblem to tackle the fractional term. Meanwhile, the computation resource allocation is derived in a closed-form expression. Simulation results confirm that for the IRS-assisted MEC system, the proposed optimization method is effective.