Joint beamforming and power allocation for intelligent reflecting surface-aided millimeter wave MIMO systems

As more challenging demands and potential applications increase in the sixth generation (6G) wireless networks, the future of mobile communications is exciting. Recently proposed intelligent reflecting surface (IRS) is a promising technology for improving the communication performance and building a programmable wireless propagation environment. However, high-precision configuration or perfect phase estimation of the IRS is unfeasible. In this paper, we study the IRS-aided millimeter wave (mmWave) multi-user multi-input multi-output (MIMO) systems with phase errors. Our objective is to maximize the achievable sum rate by joint optimizing the finited-resolution passive beamforming at the IRS, combiner, precoder, and power allocation. We first propose an iteratively sequential update (ISU) algorithm which can significantly reduce the complexity compared with the traditional exhaustive search algorithm (ESA) and cross-entropy (CE) method. Then we propose to use the genetic algorithm (GA) to further enhance the performance. The proposed algorithms can meet the requirements of low complexity and high performance respectively. Simulation results show that the IRS-aided systems can improve communication performance and coverage significantly compared with the conventional system. In addition, we observe that the system performance will not increase continuously with the number of the IRS elements when the total area of the IRS is fixed.