Successive decoding of multiple space time coded streams in multi-input multi-output systems

We propose a MIMO technique combining array processing (AP) with space-time code with antenna grouping. The transmit antennas are first grouped into subsets. Each subset employs a separate space-time code. At the receiver, after AP, each space-time coded antenna group is decoded independently. Both ZF and MMSE approaches are considered as the optimization criterion for AP. Successive cancellation is proposed to further improve the performance. The proposed algorithms are compared to various different types of MIMO schemes. When Alamouti code is employed for each data stream together with the proposed AP techniques, it is shown that at data rate equal to 4.8, 10.8, 14.4, and 21.6Mbps, the proposed methods can provide significant gains over the layered space-time (LST) architecture. For example, the 4×4 NMSE AP with Alamouti code at 21.6Mbps outperforms the 4×4 LST at 14.4Mbps by about 3 dB. Furthermore, the performance gain between the proposed algorithm and LST enlarges as the data rate increases.