Blind multiuser detection and interference cancellation in DS-CDMA mobile radio systems

This paper deals with blind adaptive multiuser detection and interference cancellation for direct sequence-CDMA wireless communication systems using antenna arrays. Such techniques have been considered as powerful methods for increasing the overall system quality, capacity and coverage. With a large number of adaptive weights, LMS type algorithms suffer from poor convergence while conventional least squares techniques can be computationally prohibitive. To mitigate these problems, this paper presents a two-stage blind adaptive receiver architecture which carries out multiuser detection using an optimal bounding ellipsoid (OBE) algorithm and direction-of-arrival estimation based beamforming using a novel, simple yet robust algorithm referred to as differential phase smoothing. The unique discerning update property of the OBE algorithms allows for a reduced complexity receiver. Furthermore, the adaptive multiuser detector also inherits improved convergence and tracking properties. Simulation results demonstrating its performance are presented.