A method of traffic congestion state detection based on mobile big data

Abstract

Urban traffic congestion state detection has been a problem of concern at home and abroad. The mainstream and traditional ways to detect traffic congestion state include manual survey, fixed traffic information collection technology and mobile traffic information collection technology. These methods all have obvious flaws such as the sample size of these methods is limited and additional equipment is needed. So the credibility of these methods' results is low and the cost is high. In recent years, mobile phones are becoming more and more common. When vehicles running on the road, the mobile phones of the people in the vehicles will connect the nearby base stations and a large number of data containing location and time information will be recorded. We call these data mobile big data. In this paper, we propose a method of road traffic congestion state detection based on mobile big data. The distributed file system HDFS is used to store data and Apache Spark is used to process data. Our method is based on huge data and no extra equipment is needed. So our method can realize road traffic congestion state detection with lower cost, shorter period and more credible results.