Research On V2V Based Traffic Congestion Detection And Results Dessemination Method

The problem of traffic congestion has become increasingly severe and has caused many adverse effects on the economy,the environment,traffic safety,and traffic efficiency.Although traditional intelligent transportation systems based on geomagnetic loop,video,and other technologies have played an important role in detecting and relieving traffic congestion.These technologies and equipment still have great limitations in terms of detection range,detection accuracy,information processing,and release.With the development of internet of vehicle,the traffic congestion detection based on vehicle-to-vehicle(V2V)technology is a new idea.Based on the above background,a V2V-based application is constructed for internet of vehicle.Compromising the merit of fuzzy logic,hypothesis testing and other methods,a traffic congestion detection method based on improved D-S evidence theory is proposed.With the existence of poor network link reliability,broadcast storms,large network load and other issues,congestion data dissemination method based on intersection is researched and proposed.The main work is as follows.(1)By the analysis of the architecture of the internet of vehicle system,a typical internet of vehicle application scenario based on V2 V is constructed,and the evaluation standard of traffic congestion is researched.According to the characteristics of the internet of vehicle,the evaluation standard of traffic congestion is defined.(2)A traffic congestion detection method based on improved D-S evidence theory is proposed.Vehicle density is estimated by vehicle-to-vehicle(V2V)interactive beacon messages.Fuzzy logic is constructed to detect local traffic congestion level by real-time vehicle speed and density.Local traffic congestion level is tested by hypothesis testing.The detection results are processed with the neighbor vehicle's congestion level message fed back by the V2 V.Then the local traffic congestion level is verified.Based on the improved D-S evidence theory,the local traffic congestion level is integrated to estimate the regional traffic congestion level.(3)A traffic congestion-oriented congestion data dissemination method is proposed.Using V2 V and sending data collection packets and data collection packet reply packets,the real-time traffic and network state is measured,and the connectivity between nodes at intersections and next-hop nodes on adjacent road segments is evaluated.Based on the results of real-time traffic and network status measurement and connectivity evaluation,a comprehensive evaluation is performed on the adjacent road sections that have connectivity with the nodes at the intersection;The results of the comprehensive evaluation of the road sections are used to select the dissemination route at the intersections.(4)A simulation platform of internet of vehicle based on Veins is built.The virtual traffic environment is constructed.The evaluation indicators and simulation parameters is defined.The proposed method for traffic congestion detection and data dissemination is evaluated on the vehicle simulation platform.The evaluation results show that compared with the detection accuracy of “loop method” and “CoTEC method”,the detection accuracy rate of the traffic congestion detection method based on improved D-S evidence theory proposed in this paper is improved by 5.5% and 7.5%;Compared with the enhanced greedy forwarding method based on traffic aware(EGyTAR),the intersection-based congestion data dissemination method proposed in this paper has greatly improved the evaluation indicators,such as packet delivery rate,network load and transmission delay.