ADMM Based Symbol-Level Precoding for MU-MISO Downlink With Low-Resolution DACs

Abstract

In this letter, we investigate the problem of symbol-level precoder design with low-resolution digital-to-analog converters (DACs) to implement wireless multiuser multiple-input single-output (MISO) downlink communications. Specifically, we aim to maximize the minimum distance between the noise-free received signal and its decision boundaries via a min-max fairness design. Due to the nondifferentiable objective and finite-alphabet constraint, the problem is NP-hard and difficult to tackle. To handle this problem, we first reduce the dimension of the variables by solving a relaxed problem. Then, the problem is transformed to a smoothed form which can be efficiently solved using the alternative direction method of multipliers (ADMM)-based framework. In addition, we propose a closed-from solution based on majorization-minimization method to provide a low-complexity near-optimal solution. We further provide the convergence and computational complexity analyses, and validate the performance by simulations.