Mobility pattern of individual user in dynamic mobile phone network using call data record

Abstract

The pervasiveness of cellular telephones, along with rapid advancement in cellular technology, wireless networks and mobile applications, has made them increasingly effective in our daily lives. Individual human activity with respect to travel behaviour is still in a less explored stage. In the past few decades, travel patterns and human activity have become more complex and exhibit a high degree of variations in space and time. In this study, how and in what extent Call Data Records (CDRs) from mobile phones provide the underlying activities of anonymised customers when a user initiates or receives a voice call or message is investigated in detail. Thus, this paper attempts to discover personal human mobility patterns where location, duration and days are the essential attributes in the interpretation of results. Profile-based paging algorithm has been proposed based on the individual human travel pattern. The finding suggests that the proposed algorithm is three times more efficient than conventional paging (blanket paging) and two times more effective than various other intelligent paging algorithms. Further, a substantial saving of about 35% in bandwidth is also obtained from the validation results.