| Code Division Multiple Access (CDMA) has proved to be an efficient and stable means of communication between a group of users which share the same physical medium. Therefore, with the rising demand for high-bandwidth multimedia services on mobile stations, it has become necessary to devise methods for more rigorous management of capacity in these systems. While one of the substantial techniques for regulating capacity in CDMA systems is through power control, the mathematical complexity of the regarding model complicates the analysis and the subsequent implementation of useful generalizations and extensions.;The major part of this thesis is the analysis of the capacity optimization problem equipped with increasing sets of constraints and utility functions. These features are incorporated into the problem in order to produce solutions deployable in practical systems. Cases of multiple-class systems are also analyzed and more accurate system models are implanted in the problem as well. Subsequently, after single-cell systems are carefully examined, one chapter is devoted to the analysis of multiple-cell systems. These systems are modeled based on an inclusive set of parameters.;For each problem, a solver is developed and experimental results are discussed. Some of the material presented in this thesis has been previously published in a number of articles, as listed in Appendix B.;In this thesis, the classical problem of capacity optimization in the reverse link of a CDMA system is analyzed. Here, it is shown that the classical formulation of the problem is solvable through examination of a finite set of transmission powers, for which closed forms are given. Although, this method leads to a more accurate and a faster solution to the classical problem, it is noted that the classical problem is very prone to yielding partial solutions in which the calculated system capacity is not realizable in a practical setting. The developed mathematical model, however, is shown to be applicable to more general definitions of the problem. |
