On the Fractional Movement–Distance Based Scheme for PCS Location Management with Selective Paging

Abstract

Abstractabstract In this paper we propose a fractional movement–distance based scheme. In our proposal, a mobile terminal stores in its local memory the identification of a set X of cells within a distance H (in terms of cells) from the cell where the last contact between the mobile terminal and the network occurred. When the mobile terminal visits a cells the following action is taken. If the identity of the visited cell is stored in the local memory of the mobile terminal, its movement counter is reset to zero. Otherwise its movement counter is increased by one unit. In this last case, if the counter has reached a given threshold d i the mobile terminal sends an update message with probability q i and with probability 1–q i the mobile terminal waits for the next decision which is taken in case of the next threshold d i + 1 being reached. A set of N movement thresholds {d i } (integer numbers) with the corresponding set of N probabilities {q i } has been used.Furthermore, for delivering incoming calls we have considered two selective paging schemes that are combined with the proposed location update scheme. The tradeoff between our proposals on location update and paging has been analyzed by means of standard Markovian tools. Then, it has been shown that in some cases where selective paging is implemented, the optimal mean value of the set {d i }, \(\overline{d}\), is a real number, not necessarily integer. This optimal value minimizes the total location management cost per call arrival (location update plus paging cost) and outperforms the movement based scheme. However, the distance based scheme still offers a better performance. But, with few memory requirements for the mobile terminal (the set X), our proposed scheme is very close to the mentioned distance based scheme.KeywordsCenter CellLocal MemoryMobile TerminalLocation UpdateIncoming CallThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.