A Gradient Ascent Based Low Complexity Rate Maximization Algorithm for Intelligent Reflecting Surface-Aided OFDM Systems

Since a typical intelligent reflecting surface (IRS) comprises a large number of passive reflecting elements, it is crucial to update the reflection coefficients using a fast optimization algorithm. In this letter, we propose a Gradient Ascent (GA) based IRS coefficient optimization algorithm, to obtain the optimal IRS phase shifts that maximizes the achievable rate of IRS-Orthogonal frequency division multiplexing (OFDM) systems. We first transform the conventional complex phase optimization problem into a real-valued maximization problem with box constraints containing real variables. This transformation enables optimization through low-complexity real valued operations. We demonstrate the efficacy of the proposed GA method over the state-of-the-art solutions in terms of convergence speed, computational complexity and achievable rate through numerical simulations.