Intelligent Reflecting Surface Enhanced Wireless Networks: Two-Timescale Beamforming Optimization

Intelligent reflecting surface (IRS) has drawn a lot of attention recently as promising new solution to achieve high spectral and energy efficiency for future wireless networks. By utilizing massive low-cost passive elements, the propagation environment becomes controllable thus can be made favorable improving communication performance. Prior works on IRS mainly rely instantaneous channel state information (I-CSI), which, however, is practically difficult obtain IRS-associated links due its operation large number elements. To overcome this difficulty, we propose in paper two-timescale (TTS) transmission protocol maximize achievable average sum-rate an IRS-aided multiuser system under general correlated Rician model. Specifically, phase shifts are first optimized based statistical CSI (S-CSI) all links, which varies much slowly compared their I-CSI; while transmit beamforming/precoding vectors at access point (AP) then designed cater I-CSI users' effective fading channels with shifts, significantly reducing training overhead beamforming design complexity over existing schemes channels. Besides, ease practical implementation, consider discrete each element IRS. For single-user case, efficient penalty dual decomposition (PDD)-based algorithm proposed, where updated parallel reduce computational time. TTS stochastic successive convex approximation (SSCA) by constructing quadratic surrogate objective function, cannot explicitly expressed closed-form. Simulation results presented validate effectiveness our proposed algorithms evaluate impact S-CSI correlation