Reduced-Rank Statistical Signal Processing in Wireless Communications

• Main Authors: Sheng-Fu Wnag, 王聖福
• Other Authors: Chia-Chang Hu
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/15895559239416046111

碩士 === 國立中正大學 === 電機工程研究所 === 93 === In direct-sequence code division multiple access (DS-CDMA) communication system, problems of both multiple access interference (MAI) and near-far effect cause performance to drop. Many multiuser detection schemes have been developed to mitigate near-far effect and suppress MAI. However, these methods usually are too complicated to implement in practical system. A considerably lower complexity version of the proposed methods that utilizes the concept of the multistage Wiener filter (MWF) introduced recently by Goldstein, Reed and Scharf. The capability of proposed methods provides the flexibility needed to compromise between the computational complexity and performance. It needs to compute orthogonal blocking matrices in multistage Wiener filter, but orthogonal blocking matrices calculation are complicated. We propose new method which using a series of nested Householder transformation in MWF implementation, not only performance can remain unchanged, and needn't calculate orthogonal blocking matrices. According to [5], we propose the multistage vector Wiener filter is generalized to the multistage matrix Wiener filter in order to apply in multiple input multiple output (MIMO) system of orthogonal transmit diversity-simulcast (OTD-S). Simulation results show that that proposed detector obtains a better performance than that of the conventional detector in all simulations. The proposed detector achieves a superior performance with lower filter bank. This makes it possible to design a lower complexity receiver without a large loss in performance in comparison with minimum mea-square error (MMSE) detector.