Comparative Study of Adaptive Multiuser Detections in Hybrid Direct-Sequence Time-Hopping Ultrawide Bandwidth Systems

Abstract

This chapter considers low-complexity detection in hybrid Direct-Sequence Time-Hopping (DS-TH) Ultrawide Bandwidth (UWB) systems. A range of Minimum Mean-Square Error (MMSE) assisted Multiuser Detection (MUD) schemes are comparatively investigated with emphasis on the low-complexity adaptive MMSE-MUDs, which are free from channel estimation. In this contribution, three types of adaptive MUDs are considered, which are derived based on the principles of Least Mean-Square (LMS), Normalized Least Mean-Square (NLMS), and Recursive Least-Square (RLS), respectively. The authors study comparatively the achievable Bit Error-Rate (BER) performance of these adaptive MUDs and of the ideal MMSE-MUD, which requires ideal knowledge about the UWB channels and the signature sequences of all active users. Both the advantages and disadvantages of the various adaptive MUDs are analyzed when communicating over indoor UWB channels modeled by the Saleh-Valenzuela (S-V) channel model. Furthermore, the complexity of the adaptive MUDs is analyzed and compared with that of the single-user RAKE receiver and also with that of the ideal MMSE-MUD. The study and simulation results show that the considered adaptive MUDs constitute feasible detection techniques for deployment in practical UWB systems. It can be shown that, with the aid of a training sequence of reasonable length, an adaptive MUD is capable of achieving a similar BER performance as the ideal MMSE-MUD while requiring a complexity that is even lower than that of a corresponding RAKE receiver.