الفهرس 
introductionOnly 14 pages are availabe for public view
 |  | from | 16 |  |  |
| | | from | 16 | | |
Abstract
The thesis aims to develop practical multiuser detectors for frequency selective fading channels. The thesis concentrates on adaptive Multiuser Detection (MUD) techniques. It focuses on DS-CDMA systems applying short spreading codes, which repeal every symbol period to take advantage of their cyclostationary properties. Use of short codes facilitates the use of advanced signal processing techniques, where adaptive receivers could learn about the interference structure and effectively suppress it. MUD techniques for multiratc DS CDMA systems supporting variable data rates are also developed. Chapter 1 introduces the issue of multiuser detection. It explains the motivation behind MUD and its promising revenues. In Chapter 2, a literature review on MUD receivers is presented. The optimal MUD and the currently used conventional detector are described. Suboptimal detectors lying between these two extremes arc discussed. Detectors arc classified into linear and nonlinear detectors. We compare the performance of some detectors through simulations. A focus is, then, made on adaptive detectors. The chapter is concluded with a comparison of the basic detector types. Chapter 3 focuses on adaptive architectures for centralized receivers. The basic system model to be employed in the rest of the thesis concerned with single rate systems is developed. Different receiver architectures are reviewed. The most popular algorithms for the adaptive implementation of the Minimum Mean Square Error (MMSE) receiver are discussed. The performance of the fully connected receiver is reviewed. We compare the performance of different architectures by various simulations. Most of the issues discussed in this chapter are considered as a basis for the rest of the thesis. In Chapter 4, the background behind blind minimum variance detectors, also called Minimum Output Energy (MOE) detectors, is laid. The basic MOE detector applicable mainly in AWGN channels is reviewed. The idea of mismatch and desired signal cancellation is explained. The rest of the chapter deals with other detectors based on the MOE detector but designed for multipath channels or to mitigate the problem of signal cancellation. A main contribution of the thesis is presented in Chapter 5. We propose a minimum variance detector with constrained surplus energy. The detector is designed for multipath