Yimeng Feng

Postdoctoral Research Associate

University of New South Wales

Dr Yimeng Feng is a Postdoctoral Research Associate exp ert in the Internet of Things (IoT), . smart farming, intelligent transportation systems and cloud computing, with a number of papers published in IEEE top journals and IEEE/ACM conference proceedings. IEEE and ACM are the two largest Computer Engineering or ganisations in research fellow position at Western Sydney University for the world. her She hold s a visiting smart farming expertise. IoT research provides maximum benefits to the DigiTech industry. patent commercialised She also own s Her a product using a lowcost magnet ic sensor to monitor on other achievements include fundings, scholarships and awards, grant, UTS showcase finalist award, collaborative such as road vehicles. Her the UTS PhD travel Research Degree Program Scholarship, Faculty of Engineering and Informati up Scholarship. on T echnology Scholarship, and UTS FEIT Female Top

Smart Road Sensing for Intelligent Transportation Systems

In recent years, intelligent transportation has attracted more and more attention. Sensing the road traffic flow and vehicle speed is one of the main focuses in intelligent transportation system. However, the current road traffic monitoring methods are not costly enough for large-scale deployment. This research is committed to investigating advanced low-cost intelligent road sensing technology for intelligent transportation system. In particular, this thesis proposes technologies on traffic flow detection, vehicle classification, and vehicle speed estimation, with the use of low-cost and lightweight magnetic sensors on the road. A novel method of vehicle speed estimation through a single magnetic sensor is first proposed. Vehicle speed estimation and vehicle classification through a magnetic sensor is then presented, based on the fluctuation of the magnetic field transmission signal. Aiming at solving the problems associated with multi-sensor clock registration and missing data, an efficient trajectory-oriented multi-hypothesis model is developed to optimize the associated processing of vehicle data by multiple magnetic field sensors. The research presented in this thesis provides efficient and low-cost solutions to smart road sensing in intelligent transportation systems.