点击次数：
论文类型：期刊论文
发表时间：2013-01-01
发表刊物：AD HOC & SENSOR WIRELESS NETWORKS
收录刊物：SCIE、Scopus
卷号：18
期号：3-4,SI
页面范围：277-292
ISSN号：1551-9899
关键字：Wireless sensor network; intelligent transportation systems; traffic data quality; cusp catastrophe theory; batch estimation filter
摘要：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.