Spatio-temporal channel identification and equalization in the presence of strong co-channel interference

We address the problem of identifying and equalizing communication channels in the presence of strong co-channel interference (CCI). In this paper, we consider the interference and noise as colored noise with unknown covariance. In addition to the finite alphabet property of the information sequence and the inherent algebraic structure of the data model, we show empirically that potential improvement in sequence detection and channel estimation accuracy can be achieved from temporal diversity where the channel outputs are observed from a tapped delay line. To exploit these underlying structures and constraints under a common framework, the proposed algorithm optimizes a weighted least-squares cost function using an iterative re-weighting alternating minization procedure. Numerical examples are presented and show that the proposed algorithm is capable of achieving reliable channel identification and equalization in the presence of strong CCI at moderate SNR.