Root Mean Square Decomposition for EST-Based Spatial Multiplexing Systems

We consider the transceiver design for multiple-input-multiple-output (MIMO) systems when the channel state information (CSI) is available at the transmitter as well as the receiver. First, we propose an open-loop low-complexity MIMO spatial multiplexing scheme based on the energy spreading transform (EST-SM). The EST-SM can spatially multiplex multiple data streams and iteratively detect the data streams with almost negligible interstream interference at sufficiently high SNR. Then, we propose a closed-loop precoding scheme suitable for the EST-SM called root mean square decomposition (RMSD) scheme. The RMSD precoding scheme combined with the EST-SM decomposes a MIMO channel into multiple subchannels with identical SNRs. This desired property minimizes bit error rate (BER) when different bit allocations on different subchannels, which cause a significant increase in system complexity, are not used. We show that when the EST-SM is used the RMSD scheme is optimal in BER performance and it achieves full diversity. Simulation results show that the RMSD scheme outperforms other existing techniques such as the geometric mean decomposition (GMD) scheme (Jiang , IEEE Trans. Signal Process., vol. 53, no. 10, pp. 3791-3803) and the uniform channel decomposition (UCD) scheme¹ (Jiang , IEEE Trans. Signal Process., vol. 53, no. 11, pp. 4283-4294) in BER performance.