We propose a geometric mean decomposition (GMD) based vector precoding (VP) for multiple input multiple output (MIMO) systems. Minimum mean square error (MMSE) criterion is used for the joint VP design. The application of GMD method eliminates the imbalance among subchannel gains and obtains a better perturbation vector than the conventional method. We then exploit the extended channel matrix for further performance improvement. Simulation results show the proposed schemes significantly outperform the existing VP schemes.

In this letter, a Tomlinson-Harashima precoding (THP) scheme is proposed for the downlink of multiuser MIMO systems with multiple antennas at each receiver. Assuming single data stream communication for each user, joint transmitter and receiver design is done to maximize the signal to noise ratio (SNR) for each user. Furthermore, a heuristic user ordering algorithm is proposed to optimize the encoding order and improve the bit error rate (BER) performance. Simulation results have shown that the proposed approach is superior to some existing precoding schemes.