Space-time blind adaptive multiuser detection in antenna array CDMA systems

In this paper, a space-time blind adaptive multiuser detection (MUD) algorithm based on signal subspace estimation is proposed in antenna array CDMA systems. The proposed method employs a new orthogonal projection approximation subspace tracking (OPAST) algorithm for adaptive subspace estimation, with a linear computation complexity and a global convergence property like the PAST algorithm, and Kalman-type algorithm for adaptive channel estimation. Moreover, unlike the PAST algorithm, the OPAST algorithm guarantees the orthogonality of the weight matrix spanned by the signal subspace at each iteration, which makes the MUD based on OPAST outperform the MUD based on PAST. The proposed method exploits an antenna array, with simultaneous space-time processing, so the performance of the receiver in multipath fading environments can be enhanced.