Channel-shortening multiuser detector for DS-CDMA systems

The multiple-access interference (MAI) encountered by direct-sequence code-division multiple access (DS-CDMA) users in a multipath environment is a key issue in the detection of mobile communication users. MAI is one of the major reasons for the degradation in performance of multiuser communication systems over single user systems. Although the optimum detection technique, maximum-likelihood sequence estimation (MLSE), is well known, its complexity is exponential with the number of users, which makes it computationally unattractive. This paper presents a multiuser detector that performs close to the optimum MLSE while drastically reducing the decoding complexity. The detection technique is based on a channel shortening algorithm and allows for group detection of a subset of the users via MLSE.