Abstract

With various added services in future wireless communication systems, the demand for fast and highly reliable data transmissions is continuously increasing. Multiple access interference (MAI) in direct-sequence code division multiple access (DS-CDMA) and constrained capacity in single-input single-output (SISO) are the major obstacles to achieve this purpose. For overcoming these limitations, the use of multiuser detection (MUD) and multiple-input multiple-output (MIMO) is necessary. In the area of error control coding, the Chase algorithm, which takes advantage of the different reliabilities of the received symbols and focuses on correcting the error in the weakest symbols, can achieve a near maximum-likelihood (ML) performance and with significant complexity reduction. In this thesis, we discuss the application of the Chase algorithm in multiuser detectors for the SISO-CDMA and MIMO-CDMA systems. In the SISO-CDMA systems, we develop a parallel interference cancellation (PIC) scheme using the Chase algorithm after the matched filters (MFs). This scheme is called Chase-PIC. Instead of optimizing the weight of each user as most of PIC detectors do, this detector provides a new optimizing approach right after the MFs. Based on the analysis of the complexity and performance over the flat and frequency-selective Rayleigh fading channels, we also propose a new Chase based multistage PIC that uses the normalized least-mean-square (NLMS) algorithm at the earlier stages and the Chase algorithm at the later stages. This proposed scheme can achieve a better performance than NLMS-PIC but with less complexity. In the MIMO-CDMA systems, we proposed a new Chase based multiuser detector over the flat Rayleigh fading channels. This detector focuses on correcting the error of the weakest detected symbol in the layered space-time multiuser detector (LAST-MUD). Compared to another Chase based detector, B-Chase, the proposed scheme has less complexity but with the tendency of achieving better performance at high signal-to-noise ratio (SNR).