Character-Aware Traffic Flow Data Quality Analysis Based on Cusp Catastrophe Theory and Wireless Sen Network
Nan Ding, Guo-Zhen Tan, Wei Zhang and Yao-Dong Wang
Given the urban traffic data automatically collected by vast amounts of traffic detectors deployed in road networks, the traffic flow data quality analysis is an important issue to intelligent transportation management system. This paper focuses on an effort to develop a character-aware data quality evaluation equation and analysis algorithm based on the state-of-the-art wireless sensor network technologies applied to the traffic monitoring system. With proximity sensor readings and data fusion, the analysis result of the data quality analysis algorithm is more accurate than single detector system. Take into account of the difference and variation of traffic flow characters in the live detection scenario, batch estimation filter is adapted, and with that the quality analysis algorithm can be self-adaptable and self-adjustable according to the traffic data detection based on the active-learning mechanism. The simulation results show that this algorithm outperforms other data quality analysis algorithms with better performance and good scalability.
Keywords: Wireless sensor network; intelligent transportation systems; traffic data quality; cusp catastrophe theory; batch estimation filter.