An artificial precoding-based method for blind separation of correlated sources in mimo fir systems

Precoding-based strategy has been demonstrated to be useful in handling blind separation of correlated sources in multi-input multi-output (MIMO) finite impulse response (FIR) systems. However, due to its heavy reliance on the precoding parameters, the knowledge of these parameters must be acquired by receivers in advance, which means extra bandwidth and transmission resources are needed. In this paper, an artificial precoding-based scheme is proposed to estimate these parameters by exploiting characteristics of both the correlated sources and the precoders. The proposed scheme allows the precoding parameters to be randomly generated, which simplifies the transmission procedure and enhances the security of the communication network. The experimental results illustrate the effectiveness of the proposed method.