The importance of location prediction is rapidly increasing with the current trend of database applications in mobile
computing environment. However, current personal communication services network could only provide currently
maintained location information of non-idle mobile terminals. Pertinent researches predict the future location based on
tangent velocity approaches, which require mobile terminals to spend lots of precious electronic power to sense and then
measure a sequence of positions for predicting the future tangent velocity, and the prediction is effective only within a short
range of time. In this study, we propose an approach to predict future locations of mobile terminals based on the moving
behaviors mined from their long-term moving history. Location prediction based on moving behavior requires no power
consumption for position measurement, and the prediction results are effective for a long time without requiring the
queried clients to be non-idle. With the help of moving behavior, we propose several location prediction operators for
location query. Finally, we demonstrate the accuracy of the location query operators through simulation statistics. The
experimental results show that the predictions are accurate enough for regular moving mobile terminals.
r 2006 Elsevier B.V. All rights reserved.
