New robust iterative minimum mean squared error‐based interference alignment algorithm

Abstract

Interference alignment (IA) is a promising technique for multiple input multiple output interference channels based systems, achieving the theoretical bound on degrees of freedom. However, these gains are reduced in the presence of imperfect channel state information (CSI), because of quantisation or channel estimation errors. In this Letter, the authors propose a new robust iterative IA minimum mean squared error-based algorithm, which includes these channel errors in the IA design. The results show that the proposed robust IA algorithm outperforms the known IA-MMSE algorithms, for low-to-moderate variance of CSI errors.