Improving location management for mobile users using two frequently visited locations table

Abstract

With the rapid growth in mobile and wireless computing technology, location management techniques become a paramount issue to enable the terminals to continuously receive services when moving from one location to another. Future Personal Communication Systems (PCS) will support a huge user's population and a network spanning over a large geographical area. In such systems, the signalling traffic and processing overhead for locating mobile users is expected to increase dramatically. In this paper, a strategy for a movement based location update procedure is proposed that would help us to optimise the cost in mobile computing. Our scheme is based on the observation and the probability of the mobile units to be in a Registration Area (RA). Depending on that probability value, the Frequently Visited Location Areas of Mobile Users are collected and stored in a data structures called FVLF-Table. Here two tables namely MFVL and LFVL are maintained which contain an optimal number of RAs, organized according to the decreasing order of probability that a Mobile User may appear in the corresponding RA. The most high order probabilities are kept in MFVL Table whereas the remaining and considerable RAs are ordered in LFVL table. This Two Table Scheme implements binary search instead of linear search in the callee list for the reduction of paging cost. Moreover this scheme also avoids the high cost of querying the callee's HLR.