RIS-Aided Cell-Free Massive MIMO System: A Two-Timescale Optimization Scheme

This letter investigates the rate performance of a two-timescale transmission scheme for reconfigurable intelligent surface (RIS)-aided cell-free massive multiple-input multiple-output (MIMO) system over spatially-correlated Rayleigh fading channels. For this system, the phase shifts of the RIS and the power control coefficients of the access point (AP) are synergistically controlled based on statistical channel state information (CSI), while the precoding of the AP is adjusted according to instantaneous aggregated channel knowledge. Firstly, we derive an expression of the downlink achievable rate with imperfect CSI and maximum ratio transmission precoding in closed-form being dependent on long-term statistics. Then, an alternating optimization algorithm based on accelerated gradient ascent and quadratic transform is proposed to solve the problem of maximizing the sum achievable rate. Finally, the simulation results confirm the accuracy of the analytical expression and the effectiveness of the proposed algorithm.