A reduced complexity detection algorithm employing adaptive LMS filters for CDMAMIMO systems using permutation spreading in slowly varying Rayleigh fading channels

Abstract

In wireless communications, research in the area of advanced detection techniques and adaptive filter algorithms for CDMA and MIMO systems in Rayleigh fading channels has been of a great interest in the past several years. On one hand, most detection algorithms can acquire satisfactory performance. On the other hand, however, the performance improvement can be achieved at the cost of an increase in computational complexity. In this thesis, we present a reduced complexity detection technique based on adaptive Least-mean-square (LMS) filtering for CDMA/MIMO systems using permutation spreading in slowly varying Rayleigh fading channels. By simulating and comparing the BER performance of 27 different models related to 4 different detections techniques (Wiener, Maximum likelihood detection (MLD), LMS, conventional method) in different Rayleigh fading channels and by analyzing and evaluating the computational complexity of Wiener, MLD, and LMS tracking detection solutions, it is not difficult to find that the proposed LMS tracking detection algorithm for CDMA/MIMO systems with permutation spreading has shown its outstanding advantages: lower complexity, without sacrificing too much BER performance, and dynamic tracking of frequency non-selective Rayleigh fading channels.