Traffic Abnormal Information Detection And Dissemination Mechanisim In Urban Expressway Vanets

Along with the rapid development of the Internet of Things,Vehicular Ad Hoc Network(VANET),as one of the important components of Intelligent Transportation System(ITS),achieves the integration of “Vehicle-Road-Human” by multiple communication methods,and develops into an information sharing and interacting platform which is oriented to computation and service.Serving as the backbone of city road networks,urban expressways are crucial for the operation of traffic systems.Once incidents occur on the heavy traffic expressways,not only will the safety of people and vehicles be threatened,but also the traffic jam will be blocked to release due to the relatively closed and complex structures,which results in a more serious chain reaction either in time or space.Therefore,it is important to reduce the negative effect of incidents on urban expressways.Information dissemination is a fundamental and important issue in VANET,it could supply drivers with the beyond-view message about other vehicles and real-time traffic information,which is of great significance to enhance traffic safety and efficiency.In this thesis,we focus on the study of traffic abnormal information detection and dissemination mechanism in urban expressway VANETs.The main contents and contributions of this thesis are summarized as follows:First,a distributed multi-layer abnormal information detection mechanism is proposed,which can deal with the limitations of existing traffic abnormal detection methods when applied to the urban expressways.This mechanism can detect the traffic anomalies in a timely manner,and determine the extent of abnormal area and severity of traffic condition by estimating and quantifying the traffic performance in real time.In this scheme,the timestamp-based averaging is firstly used to process beacons and sensor data that are periodically exchanged between vehicles for eliminating the redundant and erroneous data.Then based on the decision parameters,traffic velocity and density,vehicles can independently estimate the local traffic state with fuzzy logic.Finally,when an abnormal situation is detected,the mechanism will share the individual estimations made locally by different vehicles to collaboratively and accurately detect and characterize the road traffic performance.On that basis,the scheme develops a spatial–temporal effectiveness model to control the dissemination area and survival time of the abnormal information.Second,due to complex road structures,quality-of-service(QoS)requirements are various in different urban expressway segments,therefore a location-based warning information dissemination mechanism is proposed in this thesis.Warning messages can be quickly and reliably disseminated in the long propagation area,which utilizes the sender-oriented multi-hop broadcast mechanism.For the purpose of minimizing the propagation delay,feasible relays are selected by considering the distance to the sender and the packet reception probability.Backup vehicle transmissions are used simultaneously to ensure the delivery ratio and reduce the delay further.Moreover,when transmitted to the specific area,the messages can maintain for some time with high coverage ratio and low redundancy by the hovering-based dissemination strategy.The implementation of the proposed dissemination mechanism will provide valuable real-time information.Through this,traffic abnormal information could be obtained in time,the approaching vehicles can receive the warning and make more reasonable driving decisions,which helps to reduce the negative effects of the incidents and improve the safety and efficiency of city traffic.