Statistical pre-filtering for MIMO-OFDM systems

This paper is concerned with the application of MIMO-OFDM in wideband cellular or fixed wireless systems. We study the performance improvement for downlink, where (i) the number of base station (BS) antennas is greater than the number of mobile terminal (MT) antennas; (ii) the propagation channel is characterized by several dominated cluster paths, where the delay spread for each cluster is small, but the overall delay spread can be very large; and (iii) the transmit antennas are correlated but the receiver antennas are uncorrelated. We propose a statistical pre-fiftering (SPF) technique for spatial multiplexing MIMO-OFDM, using statistical beamforming and signal prealignment (PAL) to artificially generate the required number of uncorrelated channels to support the spatially-multiplexed data streams, and to reduce the delay spread of the channel. The required cyclic prefix length of the proposed technique is reduced considerably and therefore increasing the effective throughput. Beamforming will also provide additional array gain to the system. Results from computer simulations are presented to show the performance of the proposed transmission technique.