Abstract
Cumulant based inverse filter criteria for blind equalization of multi-input multi-output (MIMO) systems with non-Gaussian measurements can be applied to blind equalization of multiuser asynchronous DS/CDMA systems in multipath, but each pair of detected symbol sequence and estimated overall multipath channel (convolution of multipath channel and spreading code) cannot be identified with the associated active user. In this paper, a user identification algorithm (UIA) is proposed that further processes each estimated overall channel response by a bank of matched filters of user signature sequences followed by a comparator to identify each detected symbol sequence with the associated user. Then some simulation results are presented to support the efficacy of the proposed UIA. Finally, some conclusions are provided.