Implementation of Linearly Precoded MIMO Communication Systems with BSS Techniques

One of the problems of implementing precoded Multiple Input Multiple Output (MIMO) communication systems is that Channel Status Information (CSI) has to be available at the transmitting side. In practical applications, CSI can be sent through a feedback channel that is severely limited in data rate. In this paper we explore how to use Adaptive Blind Source Separation (ABSS) algorithms to detect strong channel time-variations and considerably reduce the feedback channel overhead. Moreover, ABSS algorithms also provide adequate matrices to implement the decoding process. Simulation results show that degradation in performance due to the utilization of ABSS decoding matrices and imperfect CSI is acceptable.