Adaptive blind decoding of unitary space-time constellations in ISI channels

Blind equalization and decoding of unitary space-time codes is considered for inter-symbol interference (ISI) channels. Previously, an adaptive equalization method that can equalize channel blindly was proposed, by exploiting the structure of the codes proposed by Hochwald et al. In this paper, the performance of the decoder followed by this equalizer is investigated. After giving a modified version of the algorithm in order to avoid undesired convergence, a generalized likelihood ratio test based non-coherent decoder is presented. Semi-analytic methods are developed to investigate the performance of non-coherent decoder followed by zero-forcing and minimum-mean squared-error equalizers. Simulation results indicate that the performance of the algorithm is in between ZF and MMSE equalizers.