| The overhead due to location management techniques used in the present mobile cellular systems, in particular in high traffic spots, has been a source of concern, and is expected to increase dramatically in future wireless systems. In this dissertation, the problems associated with the currently used methods of location management are studied. To this end, the various components of the location management cost, namely the paging cost, the access cost, and the signaling cost, are identified and a simple analytical framework is presented to compare various paging schemes. Based on these closed-form expressions derived for the bandwidth cost, the optimal size of the location area (LA) is found for each scheme for a given call-arrival rate and boundary-crossing rate . Intelligent paging schemes using information about the mobile speed, and most recent interaction area are proposed, and the paging cost is quantified in terms of the probability of a successful first paging step. It is shown that, the framework presented earlier can be used for analyzing intelligent paging cost, also. When compared with the existing schemes, the benefit of the intelligent scheme is found to be much more profound in the pedestrian-dominated PCS environments. The impact of random user motion on intelligent paging schemes is examined and quantified. To this end, a straight-line motion pattern and two Markov chain formulations are presented to model the random movement, of mobile users within the location area (LA). Results show that random user motion has a clear adverse impact on the cost of intelligent paging. All-step intelligent, paging is proposed, and it is shown that it can substantially reduce the paging cost compared to the other paging schemes. Finally, it is shown that if the paging cost is computed based on the number of paged cells only, in certain cases, one may obtain misleading results. Hence, a new cost function is introduced to analyze all multi-step paging procedures and it is shown that multi-step paging methods with constrained or unconstrained delay can be analyzed via a closed-form expression. |
