JultikaUniversity of Oulu repository

Abstract

A multiple-input multiple-output (MIMO) interference broadcast channel (IBC) channel is considered. There are several base stations (BSs) transmitting useful information to their own users and unwanted interference to its neighboring BS users. Our main interest is to maximize the system throughput by designing transmit precoders with weighted sum rate maximization (WSRM) objective for a multi-user (MU)-MIMO transmission. In addition, we include the quality of service (QoS) requirement in terms of guaranteed minimum rate for the users in the system. Unfortunately, the problem considered is nonconvex and known to be non-deterministic polynomial (NP) hard. Therefore, to determine the transmit precoders, we first propose a centralized precoder design by considering two closely related approaches, namely, direct signal-to-interference-plus-noise-ratio (SINR) relaxation via sequential parametric convex approximation (SPCA), and mean squared error (MSE) reformulation. In both approaches, we adopt successive convex approximation (SCA) technique to solve the nonconvex optimization problem by solving a sequence of convex subproblems. Due to the huge signaling requirements in the centralized design, we propose two different distributed precoder designs, wherein each BS determines only the relevant set of transmit precoders by exchanging minimal information among the coordinating BSs. Initially, we consider designing precoders in a decentralized manner by using alternating directions method of multipliers (ADMM), wherein each BS relaxes inter-cell interference as an optimization variable by including it in the objective.

Then, we also propose a distributed precoder design by solving the Karush-Kuhn-Tucker (KKT) expressions corresponding to the centralized problems. Numerical simulations are provided to compare different system configurations with QoS constraints for both centralized and distributed algorithms.