An information theoretic approach to blind Multiple Access Interference suppression

This paper addresses the problem of blind interference cancellation in Direct Sequence Code Division Multiple Access (DS CDMA) systems. The Maximum Likelihood (ML) criterion is used to estimate the coefficients of a linear filter that suppresses both Multiple Access Interference (MAI) and Inter-Symbol Interference (ISI). Block-iterative and adaptive algorithms based on the Space Alternating Generalized Expectation-Maximization (SAGE) approach are proposed and their performance is compared with existing Linearly Constrained Minimum Variance (LCMV) blind receivers.