Subspace tracking based blind multiuser detection: a maximum likelihood estimation approach

A blind maximum likelihood multiuser detection scheme for DS-CDMA systems is proposed, based on signal subspace tracking. Different from conventional blind detection methods, we investigate the detection problem from the perspective of user's parameter estimation. The proposed scheme uses efficient subspace tracking algorithms and yields better performance and lower computational complexity in contrast to some of the existing detection approaches. A simulation example is given to compare the proposed method and some of the methods in literature.