ADMM-Induced Low-Complexity Precoder Design Algorithm for Rate-Splitting Multiple Access

Abstract: Rate splitting multiple access (RSMA) has emerged as a promising technique in multi-antenna downlink transmis sions, which jointly encodes the private and common messages into a common stream for removing part of interference incurred by signal multiplexing and improving the spectral efficiency. Due to the decodability constraint of the common message at each user, the precoder optimization is non-convex, which lacks efficient algorithms. For facilitating the practical implementation of RSMAinfuture wireless networks, e.g., massive multiple-input multiple-output (MIMO), we propose an alternating direction method of multipliers (ADMM)-induced iterative algorithm to optimize the precoder, where the precoder is optimized for maximizing the sum rate at multiple users under the decod ability constraint of the common message. Specifically, we first employ the weighted minimum mean square error (WMMSE) algorithm to transform the non-convex problem into a sequence of convex problems, and then the ADMM algorithm is adopted to decouple the optimization variables and obtain solutions of each subproblem in closed form. Numerical results demonstrate that the proposed algorithm not only can achieve almost the same performance as existing WMMSE-based algorithms, but also has a faster convergence rate and lower computational complexity.

Index Terms: weighted minimum mean square error, non convex optimization, low-complexity design, alternating direction method of multipliers

Recommended citation: Li G, Wang C, Zhang H. ADMM-Induced Low-Complexity Precoder Design Algorithm for Rate-Splitting Multiple Access[C]//2022 IEEE 22nd International Conference on Communication Technology (ICCT). IEEE, 2022: 49-54.
Download Paper