| Location Area (LA) planning has critical impact on cellular network design, since paging and Location Update (LU) signaling (or registration signaling) must be carefully engineered. The LA size could be made as large as the service area of Mobile Switching Centre (MSC), which can minimize the registration overhead at the expense of maximizing the paging cost. On the other hand, if each cell is taken as an LA, the paging cost is at its minimum while the registration cost is at its maximum. Accordingly, the MSC service area can be partitioned into one or multiple partitions such that the total cost of paging and registration can be minimized along with the link cost between the Base Transceiver Station (BTS) and its Base Station Controller (BSC), and that between the BSC and its MSC. This study aims to find an optimal solution for the LA planning and cell-to-switch assignment problem. In this thesis, we propose an Ant Colony Optimization (ACO) technique which takes advantages of Ant Colony System (ACS) based approach to solve the problem. The ACS algorithm is implemented and analyzed. In addition, two pheromone update strategies, the Weighting Strategy, and the Proposed Strategy, are employed to enhance the performance of the ACS algorithm. Then, the effectiveness of our approach is verified by comparing the results with that of Simulated Annealing (SA) based algorithm. |
