Optimization of Location Management Cost by Mobility Pattern

Abstract

As per one survey, every mobile user possesses two mobiles in the ratio of 1:2. Out of this 75% of the mobiles belongs to the category of smart phones. In this situation, the technology also fulfills the requirements for future advanced usage. Now-a-days mobile plays a pivotal role for connecting the global matters with in the hold of our palm. The main factor which influences the availability of mobile system in the market is speed and portability. Hence the speedy retrieval is the tharaga manthra pronounced in the minds of mobile users. In the instant world, time is the major constraint for the user who relies on the performance. If the system contains intelligence the reliability level of the system also increases, otherwise it will go down even to the point of zero.In this paper, we proposed a new mobility management schemes based on mobility pattern to minimize the total cost and to balance the Location update and search Paging. The new system its main aim is to get the speedy retrieval by using mobility pattern. In the new system one mobility pattern is maintained in each and every visited cell. If the number of pattern is increased then the movement weight is reduced and the updation cost and seeking cost is also reduced. Here some cells in the network are designated as reporting cells, mobile terminals update their positions upon entering one of these reporting cells. Due to the popularity and robustness, Genetic algorithm is used to solve the reporting cells planning problem. The new system not only satisfies the requirements of the mobile environment but also fulfills the pervasive environment, because it integrates the concepts of mobile and intelligence. By use of this intelligence, the extraction of output and its level of accuracy are very high. Here intelligence is used to identify the shortest path. The main drawback here is same time taken for first call and maintain less time for subsequent calls only. The performance of the new system can be tested by generating random data sets for number of generations 500 and 1,000, the network size taken is 4X4. Here we have evaluated the performance of the new system by comparing with some existing systems like POFLA, UPBLA and MIPN. The performance can be measured by call-to-mobility ratio, locating time ratio, the update cost ratio, time-data routed ratio. Comparatively the new system is better than any other existing system we have mentioned.