Exploiting non-Gaussianity in blind identification and equalisation of MIMO FIR channels

Abstract

The problem of blind identification and equalisation (BIE) of finite impulse response (FIR) channels in multiuser digital communications is investigated. The non-Gaussian nature and statistical independence of the users' data streams is exploited by resorting to blind signal separation (BSS) based on higher-order statistics (HOS). Two such techniques are put forward. The first technique is composed of an extension to the multiuser case of a second-order BIE method, followed by a BSS-based space-equalisation step. The second technique achieves joint space–time equalisation through the direct application of a HOS-based BSS method followed by a blind identification algorithm. In a number of numerical experiments, the first procedure proves less costly and more effective for short data records. Despite their computational complexity, interesting features such as constellation-independent channel identification and symbol recovery, and robustness to ill-conditioned channels in high SNR environments render HOS-BSS based BIE methods an effective alternative to BIE techniques exploiting other spatio-temporal structures.