Maximum-Likelihood Transceiver Design for RIS-Assisted MIMO Communications

This letter investigates the maximum-likelihood transceiver design for a reconfigurable intelligent surface assisted multiple-input multiple-output system. The goal is to jointly optimize the precoder and the reflector such that the minimum distance between the received signals can be maximized. The design problem is challenging and we propose an efficient approach to overcome the difficulty. That is, the so-called 2-D precoder is first applied at the source, and then a gradient ascent algorithm is developed to optimize the reflector. It is demonstrated that the use of the 2-D precoder can significantly reduce the computational complexity of the proposed reflector design. Simulation results show that our transceiver effectively improves the symbol-vector error rate performance with an affordable computational cost.