Impact of Imperfections on Detectors for Interference Suppression

In this paper, we carry out the performance analysis of recently proposed low complexity max log MAP detector for interference suppression under the realistic imperfections of correlation and non Gaussian alphabets. We also incorporate MMSE detector in our analysis taking into account non Gaussian alphabets however intricacy in the analysis of correlated case restricts our analysis to the case of iid fading. Employing moment generating function(MGF)-based approach, we derive upper bounds of coded pairwise error probability (PEP) and study the effect of non Gaussian interference and correlation. The novelty of this contribution is the performance analysis taking into account discrete alphabets as the existing work is based on the unrealistic Gaussian assumption for these alphabets.