A COMPARATIVE APPROACH FOR LOCATION MANAGEMENT USING ARMA, NONLINEAR AUTOREGRESSION AND NEURAL NETWORK FOR CELLULAR NETWORKS
Due to the randomness in user’s mobility, high degree of complexity arises to predict the exact location of the user and to forward a call within fixed time constraint. While searching the user, certain resources are utilized by network, which amount in cost. Resources could be bandwidth, memory of database or any form of energy consumed during the operation of locating the user. Therefore, it may be possible to derive a model based on the user’s movement pattern that can be used to suggest the likelihood of the future locations, underlying between two specified time periods. This paper presents a time series model based on regression analysis for easy tracing and better location management of the user.
cellular network, location management, static location management, dynamic location management, paging, location update, location area, mobility models.