Estimating instant traffic information by identifying handover patterns of UMTS signals

With the growing popularity of cellular phones, using cellular signaling data to collect traffic information has become an attractive technique to replace VDs (Vehicle Detectors). However, most previous studies have focused on GSM network instead of the ongoing UMTS due to some special properties in UMTS. In this paper, a passive framework with multiple handover pattern exploring and traffic information estimating strategy is proposed to fit characteristics of UMTS signaling data. Experiments from real-time UMTS signals result in a mean relative error around 14% to the information collected from VDs.