الفهرس | Only 14 pages are availabe for public view |
Abstract Managing location information of mobile terminals is an important issue in mobile computing systems. Unfortunately, current standards of location management techniques, such as GSM and IS-41, behave inefficiently for mobile terminals moving between the same set of areas, or users tending to call and move in a fzxed pattern. This is in turn results in a high location management cost. The two main operations of location management are Location update (LV) and Paging operations. There is a trade-off between the costs for location update and paging. Several approaches have been proposed to tackle with this problem. One of the most efficient techniques is to replicate user profile according to the most frequently visited location areas. To the best of our knowledge, none of the location management techniques utilizing replicated databases has reduced both location update cost and paging cost. In this thesis, we propose a new location management methodology based on replicated databases to reduce the cost of both of the two location management operations. The proposed methodology makes use of the mobile station to build a user profile carrying information of the most Frequently Visited Location Area (FVLA). Henceforward, these location areas can be used as replication placement to enhance the performance of our methodology, we have classified users according to their mobility behaviors into three classes; low-mobility, medium-mobility, and high-mobility. We compare the performance of our methodology with that of GSM and other replicated databases location management techniques in respect of communication cost, database cost, and total cost. Simulation results show that the proposed methodology performs better than other techniques. |