Performance investigation of MMSE-based interference alignment for multiuser MIMO UWB

Abstract

To achieve the full multiplexing gains or the degrees of freedom of the K-multiuser MIMO interference channel, the interference alignment technique has been proposed and investigated. The main idea of interference alignment is to design the pre-coder at each transmitter, so that each receiver can align all interference from undesired transmitters in interference subspaces by exploiting knowledge of the channel state information (CSI) of all transmission links. Currently, a closed-form solution for the pre-coding filters exists only for the case of up to three users and provides poor sum-rate performance compared to iterative algorithms. Furthermore, all available algorithms for interference alignment were investigated only for the case, where each transmitter sends only one data stream. In this paper, the MMSE-based interference alignment is exploited to improve the BER and sum-rate performance of the system especially in the low SNR regime. This paper focuses on a multiuser MIMO UWB system based on MB-OFDM with arbitrary number of independent data streams per user. The simulation results confirm that the MMSE-based interference alignment provides sum-rate and BER superior to Orthogonal Frequency-Division Multiple Access (OFDMA) at marginal performance degradation.