Iterative algorithm for uniform page traffic reduction in a cellular system

Page messages are used to locate a mobile when a call is to be terminated to the mobile. There has been considerable amount of work on minimizing the total amount of paging in the system by creating paging zones that are used to restrict the number of cells that are paged. Such minimization is good from the mobile switching center (MSC) perspective. However, this ignores the constraints on each of the individual cells. Experiments show that in many cases reducing the total amount of paging gives scarcely any relief to the most highly loaded cell. This is because such a cell ends up in too many zones and therefore receives too many page messages. In this work we study the problem of devising zones that reduce the paging on the worst-case cell. We have devised a simple algorithm that works in two stages. In the first stage we employ a traditional scheme to reduce the total amount of paging in the system. We then perform a second stage of optimization that iteratively attempts to reduce the paging traffic on the worst-case cells by removing them from selected zones in a manner so as to not increase the paging traffic in other cells. We have performed extensive experiments for this iterative scheme on the data collected from several systems. Our experiment shows that the page traffic in a bottleneck cell could be reduced by as much as 16% over and above the global optimization algorithm by using our algorithm