Soft-input soft-output multiuser detection for coded wireless multiuser systems

Abstract

Multiuser detection permits the joint detection of signals from multiple sources. An optimal multiuser detector for coded multiuser systems is usually practically infeasible due to the associated complexity. An iterative receiver consisting of a soft-input soft-output (SISO) multiuser detector and a bank of SISO single user decoders can provide a system performance which approaches to that of the single user system after many iterations. On the other hand, it has a feasible computational complexity. We first propose and analyze three types of SISO multiuser detectors, which are mainly based on the underlying ideas of the traditional (non-SISO) multiuser detectors. These are: decorrelators, linear minimum mean square error (MMSE) detectors and parallel decision feedback detectors. In contrast to these traditional detectors, our SISO multiuser detectors make good use of soft inputs provided by SISO single user decoders and in turn, provide soft outputs to them. The resulting system performance converges quickly, while keeping the computational complexity of the SISO detectors proportional to the number of users cubed. This is a significant complexity reduction in comparison to the optimal detector which has a complexity exponentially proportional to the number of users. Secondly, we consider adaptive SISO multiuser detection. As it is well known, the optimum and many suboptimum SISO multiuser detectors require a lot of a priori information of the multiuser system, such as all users' transmitted waveforms, relative delays as well as the channel impulse response. In this thesis, we apply adaptive algorithms in the SISO multiuser detector in order to avoid the need for this a priori information. We propose two adaptive SISO parallel decision feedback detectors based on the normalized least mean square (NLMS) and recursive least squares (RLS) algorithms for both synchronous and asynchronous direct-sequence code-division multiple-access (DS-CDMA) systems. Compared with traditional non-SISO adaptive detectors, our SISO adaptive detectors effectively exploit the a priori information of coded symbols to further improve their convergence performance. This a priori information is obtained from a bank of single user decoders. Monte-Carlo simulation results are presented and compared. Moreover, we extend adaptive SISO parallel decision feedback detection to asynchronous DS-CDMA systems over slow frequency non-selective Rayleigh fading channels.