In cellular CDMA systems, adjacent cells are not isolated in time or frequency. Therefore, the total interference at any cell site consists of in-cell and out-of-cell components. Existing estimations of CDMA network capacities assume that all cells are identical and that users are uniformly distributed in all cell. In practice, this condition is not true, and the cell-to-cell variations in user densities and propagation conditions will greatly affect the system capacity.

This thesis investigates possible schemes to account for topographical and user density variations in designing cellular CDMA systems. Algorithms willdeveloped to determine the best location of base stations, ones that match the distribution of users throughout the service area. Each base station has sectored antennas. We are investigating the possibility of adjusting the width (and gain) of each sector according to the density of users and the propagation conditions. The selection of base station locations and the design of sectored antennas will be optimized to improve the average and variance of the Signal to Interference Ratio (SIR). The optimized solutions will be evaluated using typical Quality of Service (QoS) criteria such as: the blocking and droping probabilities.