J4 ›› 2013, Vol. 40 ›› Issue (2): 207-212.doi: 10.3969/j.issn.1001-2400.2013.02.033

• Original Articles • Previous Articles    

Block diagonalized MMSE GMD vector precoding

GENG Xuan1;SUN Zuolei1;LIU Feng1;MA Yinghong2   

  1. (1. College of Information Engineering, Shanghai Maritime Univ., Shanghai  201306, China;
    2. State Key Lab. of Integrated Service Networks, Xidian Univ., Xi'an  710071, China)
  • Received:2012-07-05 Online:2013-04-20 Published:2013-05-22
  • Contact: GENG Xuan1 E-mail:xuangeng@shmtu.edu.cn

Abstract:

Based on the block diagonalization (BD) for the multiuser Multiple-input Multiple-output (MIMO) downlink channel, a vector precoding (VP) combined with geometric mean decomposition (GMD) based on the minimum mean square error (MMSE) criterion is proposed. The multiuser multi-antennas channel is decomposed into parallel effective sub-channels, and then the mean square error (MSE) of the transceiver signal is obtained based on the effective sub-channel. The VP combined with GMD is derived based on the MMSE criterion, and the perturbation vector contains the perturbation vector expanding constellation and residual interference vector. The MSE expression is then achieved based on the derivation results, and the realization block diagram is concluded finally. Simulation results show that the proposed algorithm outperforms the conventional BD method and its expanded method in bit error rate (BER).

Key words: MIMO system, mean square error, vector precoding, block diagonalization