Low Complexity Iterative Equalization for Severe Time Dispersive MIMO Channels

In this work, reduced complexity minimum mean squared error (MMSE) iterative equalization methods for a multiple-input multiple-output (MIMO) frequency selective channel are proposed. To combat the severe time dispersiveness of the channel OFDM is incorporated. Use of OFDM simplifies the challenging problem of equalization in a MIMO system, due to both inter-symbol- interference (ISI) and co-antenna interference (CAI). The iterative algorithms work in two stages. The first stage estimates the transmitted symbols using a low complexity MMSE equalizer, which accounts for the variance of the already estimated symbols. Then, the second stage finds the a posteriori probabilities of the estimated symbols to find their means and variances to use in the MMSE equalizer in the following iteration. Simulation results show the performance of the proposed iterative algorithm is better than the MMSE equalizer and close to the matched filter bound (MFB) at low computational cost.