A 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. The 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 will allow for group detection of a subset of the users via the MLSE.
