Future Location Prediction in Wireless Network Based on Spatiotemporal Data Mining

Handover latency is the time interval during which a mobile node cannot send or receive any packets. How to improve the latency is one of the major problems in Mobile Internet Protocol version 6 (MIPv6). One of the solutions for reducing the handover latency that has attracted various research interests is to predict mobility of mobile node. Several approaches to prediction have been proposed such as Hidden Markov models, machine learning, data mining, and so on. The approach in data mining investigates the log file of node mobility history to predict the next move of mobile node. However, such conventional studies do not consider simultaneously spatial and temporal attributes of data. In the context of wireless network, the spatial attributes of a mobile node are changing over time, therefore time constraints between locations play an important role in considering its mobility. This paper proposes a data mining based approach that utilizes spatio-temporal attributes to predict the movement of mobile node.