Mining Call and Mobility Data to Improve Paging Efficiency in Cellular Networks

Locating mobile users and devices efficiently is a critical operation in cellular networks. This is done using a combination of location update (by the mobile) and paging (by the network). The paging scheme determines how and where to search for a mobile user given the latest location update information from that user. In this paper, we consider how to increase the efficiency of the paging scheme. Much previous work has relied on simulation or modeling to design and evaluate the performance of proposed paging schemes. We take a different, data-driven approach in how we design and evaluate our solution. Specifically, we mine more than 300 million call records from a large cellular operator to characterize user mobility and create mobility profiles. We then develop a family of profile-based paging techniques, considering both static schemes and dynamic schemes which adapt as user profiles continuously get updated. We find that our paging techniques can dramatically reduce signaling load (up to 80%) with minimal increase in paging delay (usually less than 10%)