Design and performance analysis of power -controlled CDMA wireless networks with linear receivers and antenna arrays
Abstract
The mobile wireless communication environment provides several unique challenges. On one hand, network resources are limited. On the other hand, essentially all user/network parameters are time-varying. The main theme of this thesis is to understand the impact of time-varying parameters on the system performance, and to design robust systems that efficiently utilize network resources. In particular, this thesis has solved several challenging problems in the performance analysis of code-division multiple access (CDMA) networks. First we study the output multiple-access interference (MAI) distributions of NIMSE receivers in DS-CDMA systems with random spreading. This solves an important open problem. We establish that for almost every realization of the signatures and received powers, the conditional distribution of the output MAI converges to the same Gaussian distribution as in the unconditional case. We then generalize the above results to the more general case where the signatures are complex, and establish the Gaussianity of the output interference conditioned on signatures, received powers, and delays. Our results reveal that the MMSE receiver is robust. Next we take a network perspective and characterize the network capacity of various systems by using large-system analysis. We show that the matched filter receiver is sensitive to the randomness of the signatures. In contrast, the MMSE receiver is much more robust. We also generalize the results to multiple-class systems. Finally, we consider a resource allocation problem in a multiple time-scale fading environment. We use power control to compensate for large-scale fading and exploit spatial diversity to combat small-scale fading. We characterize the system performance and our results indicate that the gains of power control and diversity combining are additive (in decibels). We also characterize the Pareto-optimal (minimum) transmission power allocation.
Degree
Ph.D.
Advisors
Chong, Purdue University.
Subject Area
Electrical engineering|Systems design
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.