A Novel Freeway Traffic Speed Estimation Model with Massive Cellular Signaling Data

A Novel Freeway Traffic Speed Estimation Model with Massive Cellular Signaling Data

Tongyu Zhu (State Key Lab of Software Development Environment, Beihang University, Beijing, China), Zhixin Song (State Key Lab of Software Development Environment, Beihang University, Beijing, China), Dongdong Wu (Beijing Transportation Information Center, Beijing, China) and Jianjun Yu (Computer Network Information Center, Chinese Academy of Sciences, Beijing, China)
Copyright: © 2016 |Pages: 19
DOI: 10.4018/IJWSR.2016010105
OnDemand PDF Download:
No Current Special Offers


With the growing popularity of cell phones, using massive cellular signaling data as probe to track the vehicles movement trajectory and obtain the real-time traffic condition has become one of the most attractive candidate techniques. However, traditional approaches may offer a poor performance in removing noisy data and minimizing deviation of traffic speed in adjacent time intervals. In this paper, a novel approach is proposed to solve these two issues. The authors move noisy data by comparing the cellular signaling data with the trained data set, and adopt a modified Kalman filter algorithm to minimize the deviations. The experiment results show that the accuracy of the approach performs better in comparison to other two traffic speed estimation approaches.
Article Preview

1. Introduction

ATIS (Advanced Traveler Information System) that provides road-based traffic information to travelers is one of the important components of ITS (Intelligent Traffic System). To collect the traffic condition, many dedicated sensors on roads (such as inductive loop vehicle detector, traffic camera, microwave and supersonic detector, etc.) have been deployed in recent years. The usage of traditional on-road sensors (e.g. inductive loops) for collecting data is necessary but not sufficient because of their limited coverage and expensive costs of implementation and maintenance. In the past few years, research activities have begun to explore the alternative approaches with new data sources obtained from “in-vehicle” devices through GPS (Global Positioning System) or mobile phones. Especially the methods based on the vehicle location (i.e., Floating Car Data) which are cost-effective solutions to cope with some limitations from fixed detectors. Although the vehicles such as taxies, buses and trucks equipped with GPS have been used to collect traffic data, obtaining reliable pictures of the traffic situation on freeway is still a problem because most private cars may not be willing to sharing their private location information (Florida Department of Transportation, 2007; Guillaume Leduc, 2008; Schneider et al. 2005).

Using mobile phone as probe to track the vehicles movement trajectory, and thus obtain the real-time traffic flow has become an effective manner (Calabrese et al., 2011; Lin et al., 2011). Because of its low cost and widespread distribution, estimating traffic condition based on cellular phone signaling data has become an alternative and viable way (Bar-Gera, 2007; Wunnava et al., 2007). Prior work in this area has primarily focused on the following basic phases: Location data collection, Terminal classification, Map matching, Route determination and Traffic state calculation (David Gundlegard et al., 2009).

There are several ways using mobile phone as probe to collect traffic data. GPS enabled smart phones can provide accurate enough position data, speed and heading information. Based on these data, map matching and traffic estimation are not big problems, whereas the communication cost and privacy are still issues. The network-based probe system employs network monitoring methods that make use of cellular network signaling information, e.q., the handover measurements (Tao, S et al., 2012) or Cell Dwell Time (Suwat Pattaramalai et al., 2009; Orlik, P. V. et al., 1998).

The positioning techniques from cellular phone signaling data generally provide less accuracy than the GPS (Wunnavaet al., 2007; Quddus et al. 2007), but there are abundant cellular phones in wide-area, which will provide the position data that makes this positioning techniques very appealing (Calabrese et al., 2011). Nevertheless, accuracy has remained to be a great challenge to traffic speed estimation by cellular phone signaling data.

The inaccuracy of traffic speed estimation mainly results from two reasons. The first reason is that cellular phones from different carriers such as pedestrians by the freeway, passengers on bus and drivers in car on the freeway, bicycle riders near the freeway can all generate signaling data (Prashanth Mohan et al., 2008). The signaling data from cellular phones carried by people who don’t drive or sit in vehicles on freeway are called noisy data, and they often result in inaccuracy during estimating traffic speed and should be removed from the overall data set. However, motor vehicle carrier and non-motor vehicle carrier cannot be distinguished directly because there is no parameter in signaling data to differentiate the carriers situation (Puntumaponet al., 2008). The second reason is the number of collected signals from motor vehicles changes seriously in adjacent time intervals, increasing traffic speed deviations.

Complete Article List

Search this Journal:
Open Access Articles
Volume 19: 4 Issues (2022): Forthcoming, Available for Pre-Order
Volume 18: 4 Issues (2021): 3 Released, 1 Forthcoming
Volume 17: 4 Issues (2020)
Volume 16: 4 Issues (2019)
Volume 15: 4 Issues (2018)
Volume 14: 4 Issues (2017)
Volume 13: 4 Issues (2016)
Volume 12: 4 Issues (2015)
Volume 11: 4 Issues (2014)
Volume 10: 4 Issues (2013)
Volume 9: 4 Issues (2012)
Volume 8: 4 Issues (2011)
Volume 7: 4 Issues (2010)
Volume 6: 4 Issues (2009)
Volume 5: 4 Issues (2008)
Volume 4: 4 Issues (2007)
Volume 3: 4 Issues (2006)
Volume 2: 4 Issues (2005)
Volume 1: 4 Issues (2004)
View Complete Journal Contents Listing