Optimal MU-MIMO precoder with MISO decomposition approach

Abstract

This paper proposes a new iterative implementation algorithm for sum-rate maximization in a Multiuser MIMO system (MU-MIMO). The proposed algorithm is based on joint precoder and decoder optimization. For that we considered the best existing precoder design algorithm for a MISO multiuser system proposed in [1]. This algorithm is based on the Lagrangian sum-rate maximization procedure. For the receiving part, an optimal receiver is designed based on the system throughput maximization derived in the general case from the sum-rate expression given for a MU-MIMO broadcast channel. To link these two optimal algorithms, we use an iterative procedure transforming the MU-MIMO channel for each iteration into a MU-MISO channel trough virtual channel calculations. Finally to validate our prosed solution we compare it with an existing MMSE based iterative optimization algorithm. This algorithm proposed in [2] is based on an MMSE as well at the transmission side than at the receiving side. The obtained results demonstrate significant gains without introducing neither supplementary complexity nor resource needs.