Dissertation > Excellent graduate degree dissertation topics show

Rearch of Blind Multiuser Detection Technique in CDMA Communication System

Author: DanDan
Tutor: YuYinHui
School: Jilin University
Course: Communication and Information System
Keywords: Blind Multiuser Detection Subspace tracking Minimum mean square Independent Component Analysis Newton iteration Wavelet transform
CLC: TN929.533
Type: Master's thesis
Year: 2011
Downloads: 89
Quote: 0
Read: Download Dissertation

Abstract

Code Division Multiple Access (CDMA) system due to the soft capacity, soft switch, high spectrum efficiency and anti-interference ability in the third generation mobile communication systems (3G) wide range of applications. In a CDMA system, a plurality of user signals is to rely on the type of spread codes to be distinguished from each other, but in practical applications, the spread code is impossible to achieve the ideal state completely orthogonal, therefore when multiple users to simultaneously communicate, they will certainly have multiple access interference (MAI), with the increase in the number of users of the system, such interference is bound to become a major factor to affect the system performance and limitations of the system capacity. How to be better able to eliminate multiple access interference has become a focus for researchers in recent years. Multi-user detection technology can effectively eliminate the multiple access interference and increase system capacity to solve the problems of distance effect, which is listed as one of the key technologies of the third generation mobile communication system. Its main purpose is to detect multiple CDMA signals from interfering with each other, the signal of a particular user, fundamentally eliminate the presence of multiple access interference. Blind Multiuser Detection technology as a direct use of user data to be tested and expectations of the user information blind adaptive detection method in recent years has been the attention of many scholars and research focus of the entire field of communications. This paper around blind multiuser detection algorithm for CDMA systems expand, the first systematic introduction to the basic principles and classification of multi-user detection algorithm, and several linear blind multiuser detection algorithm is more detailed. On this basis, for the compression projection approximation subspace tracking algorithm is not strong orthogonal shortcomings, introduced the subspace tracking blind multiuser detection algorithm based on orthogonal compression projection approximation (OPASTd), simulation results show The algorithm is a better performance of the subspace tracking algorithm. For adaptive LMS algorithm convergence speed and precision by the step size greater impact on the shortcomings, an improved variable step size LMS blind multiuser detection algorithm by simulation results show that the algorithm ability to adapt to the system , and has a higher average output time signal to interference ratio. And then introduces the basic principles and mathematical model of independent component analysis (ICA), and focus on several common FastICA algorithm and its algorithm flow. The FastICA algorithm convergence speed, on the basis of the sixth-order convergence of Newton iteration, an improved fast independent component analysis algorithms, and simulation experiments to verify the superiority of the algorithm. Fast independent component analysis and improved fast independent component analysis ignore the noise estimation error, the wavelet transform is introduced into the improved FastICA algorithm and proposed a wavelet transform and improved FastICA algorithm combining ( WICA), experimental simulation show the superiority of the algorithm. By discussing the feasibility of independent component analysis applied to the CDMA system, the WICA as part of the post-processing of the two improved blind multiuser detector, and through two sets of simulation experiments show that the WICA-based algorithms consider noise than the traditional The FastICA algorithm have better performance, especially in the case of low signal-to-noise ratio, and large number of users, this is based on the WICA the blind multiuser detection algorithm has better detection performance. In summary, this study: (1) the compression projection approximation subspace tracking algorithm based on the introduced a subspace tracking based on OPASTd the blind multi-user detection algorithm. (2) on the basis of the adaptive LMS algorithm, an improved variable step size LMS blind multiuser detection algorithm. (3) for the sixth-order the Newton iterative correction format, an improved fast independent component analysis algorithm. (4) wavelet transform is introduced into the improved FastICA algorithm is proposed blind multiuser detection algorithm based on wavelet transform and improved FastICA combination.

Related Dissertations

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Wireless communications > Mobile Communications > Cellular mobile communications systems (mobile phones, mobile phone handsets ) > Code Division Multiple Access (CDMA) mobile communication
© 2012 www.DissertationTopic.Net Mobile