Sum Rate Maximization for Multiuser MISO Downlink with Intelligent Reflecting Surface

Abstract: The intelligent reflecting surface (IRS) is a planar array with massive reconfigurable passive elements, which can align the reflecting signals at the receivers via controlling the phase shifts at each element independently. Since IRS can be implemented without RF chain, it is seen as a cost-effective solution for boosting the spectral efficiency for the future communication system. This work considers the IRS-aided multiuser multi-input single-output downlink transmission, and our goal is to maximize the sum rate by jointly optimizing the transmit beamformer at the base station and the phase shifts, which can be either continuous or discrete, at the IRS. This sum rate maximization (SRM) problem is challenging, especially for the discrete-phase case. To tackle it, we first derive a more tractable equivalent formulation of the SRM problem with structured convex constraints and a smooth objective. From there, a custom-derived block-coordinated accelerated projected gradient algorithm is developed. Simulation results demonstrate that the proposed design outperforms state-of-the-art design in the sum rate performance.