A Reactive Tabu Search Based Equalizer for Severely Delay-Spread UWB MIMO-ISI Channels

We propose a novel equalizer for ultra wideband (UWB) multiple-input multiple-output (MIMO) channels characterized by severe delay spreads. The proposed equalizer is based on reactive tabu search (RTS), which is a heuristic originally designed to obtain approximate solutions to combinatorial optimization problems. The proposed RTS equalizer is shown to perform increasingly better for increasing number of multipath components (MPC), and achieve near maximum likelihood (ML) performance for large number of MPCs at a much less complexity than that of the ML detector. The proposed RTS equalizer is shown to perform close to within 0.4 dB of single-input multiple-output AWGN performance at 10^-3 uncoded BER on a severely delay-spread UWB MIMO channel with 48 equal-energy MPCs.