【摘要】As a commonly used wireless technology for short distance data exchange, Bluetooth can be integrated into Intelligent Transportation Systems to enable better and more effective traffic monitoring and management, hence reducing traffic congestion. While a Bluetooth-equipped device travels along a road, its Media Access Control address, with detection times, could be recorded anonymously with a certain probability. In this piece of work, we seek the practical objective of recovering actual traffic counts from Bluetooth data as a statistical calibration problem. A framework based on point processes is proposed to characterise the data generating processes of Bluetooth data and traffic counts. A Poisson regression model with time-varying coefficients is then developed to approximate the underlying point processes and facilitate the statistical inference. Finally, we show that statistical calibration can be used to recover the traffic counts from the counts of Bluetooth devices with uncertainty quantification effectively.

【个人介绍】Dr. Xun Xiao is currently a Lecturer in Statistics at the Dept. of Mathematics and Statistics, University of Otago, New Zealand. He received B.Sc. in Statistics from the University of Science and Technology of China in 2011 and Ph.D. degree under the supervision of Prof. Min Xie from the Dept. of Systems Engineering and Engineering Management at City University of Hong Kong in 2016. His research focuses on various statistics and data analytics problems arising from interdisciplinary areas, including industrial engineering, natural hazard, veterinary science, food science, etc. He has published more than 20 papers in peer-reviewed journals.