Impact of interference correlation on the decoding error statistics

Scenarios where interference is correlated with useful signals are often met in communications systems. Interestingly, even though uncorrelated interference mitigation has been extensively studied by the literature, there is still lack of insight into how correlation can affects information transfer processes. In this paper we formally address this issue and show how a positive correlation between the interference and useful signal power introduces a non-intuitive increase in the diversity gain, improving the error statistics with respect to the uncorrelated case. In contrast, the diversity gain decreases under negative correlation, degrading the system performance. These findings are confirmed by simulations for scenarios of practical relevance.