Energy Efficient Intelligent Reflecting Surface Assisted Terahertz Communications

This paper investigates the energy efficiency optimisation problem for an intelligent reflecting surface (IRS) assisted Terahertz communication system. In this system, a base station (BS) with multiple antennas and an IRS with a large number of reflecting elements are deployed to serve multiple users. An energy-efficient design is developed to maximise the energy efficiency of the system by considering both transmit power and IRS phase shift constraints. Specifically, we propose a covariance matrix adaptation evolution strategy (CMA-ES) and Dinkelbach's method based energy efficiency optimisation algorithm to tackle the joint optimisation problem of IRS phase shift and precoding matrix while satisfying the maximum transmit power constraint. Simulation results show that our proposed algorithm outperforms three baseline algorithms in the literature, including random selection (RS) method, local search (LS) method and cross-entropy (CE) method, in terms of much higher energy efficiency under different parameter settings.