Adaptive analytical MMA with time-varying MIMO channels and diversity in interception context
Steredenn Daumont, Daniel Le Guennec · 2009
This paper deals with blind source separation (BSS) with time-varying MIMO (multiple input multiple output) channels in interception and diversity context (Alamouti code). A new algorithm based on an adaptive analytical multimodulus algorithm (MMA) is proposed and called adaptive-AMMA. An analytical method, which converges quickly, is necessary with time varying channels. Performances obtained with adaptive-AMMA are better than adaptive-ACMA (analytical constant modulus algorithm) and simple MMA for time-varying channels. In addition, this proposed method has done jointly source separation and carrier phase recovery contrary to the adaptive-ACMA. After separation by BSS, the Alamouti code synchronization is found again and then an exploitation of the Alamouti diversity is proposed to raise sources order indetermination inherent with BSS.