A Polynomial Matrix SVD Approach for Time Domain Broadband Beamforming in MIMO-OFDM Systems

Singular value decomposition (SVD) is a useful technique to mitigate co-space interference (CSI) in multiple input multiple output (MIMO) systems employing orthogonal frequency division multiplexing (OFDM) scheme. The SVD based frequency-domain broadband beamforming (FBBF) approach uses the beamforming matrices in both transmitter and receiver sides for each subchannel matrix. However, when the MIMO- OFDM system has a large number of subchannels, the FBBF technique is very complex due to computing the SVD of each subchannel separately. In this paper, we propose a new approach to eliminate CSI and intersymbol interference (ISI) in time- domain and frequency-domain, respectively. By the use of polynomial matrix SVD based on SBR2 algorithm, the MIMO- OFDM system is decomposed to the parallel decoupled single input single output (SISO) OFDM systems. In this way, the CSI is eliminated by time-domain broadband beamforming (TBBF) in both transmitter and receiver sides and ISI is mitigated in each SISO-OFDM system by frequency-domain equalization. The performance of the proposed method is evaluated by computer simulations for fixed broadband wireless communication system developed based on the WiMax. Simulation results show that the proposed method achieves good performance in concerned SNR interval.