Research On Technologies Of Traffic Route Guidance Based On Incomplete Data In Vehicular Ad Hoc Networks

With the rapid growth of vehicles in urban traffic, road congestion has became aproblem that can not be ignored. Traffic route guidance system releases the trafficpressure in some degree. At the same time, it provides faster routes to users. It is animportant application of Intelligent Transportation System. The system collectstraffic information through Vehicular Ad-hoc Networks, processes the informationand offers navigation results to vehicles. Until now, proposed traffic routing guidancealgorithms hardly deal with incomplete traffic information, user satisfactionoptimization and fairness problem between the drivers. The related works make onlya few researches on these three aspects. Therefore, we focus on these fields.This subject is divided into three main parts according to the processing andapplication of traffic data flow. Those are incomplete traffic information recovery,trade-off between traffic load and user satisfaction, and fairness in multi-timeguidance. The first one belongs to data processing, and the last two belong to trafficroute guidance application. AutorRgression model based on Time Dimension, whichis named ARTD for short, is used to recovery the traffic informaiton thatdemonstrates the traffic flow, road density, average vehicle speed and so on. In thesecond part, we propose a method name BPR-US algorithm based on thebackpressure theory. This algorithm can achieve the balance between the traffic loadand user satisfaction. As for third part, PFU-CFD algorithm which is based on theincentive theory is applied to realize the fainess between the users. This algorithmachieves the fairness through charging to the users who benefit and compensating tousers who sacrifice.We have carried out experiments to evaluate the performance of ARTD in Cprogramming language, and simulated the Singapore expressway to verify theperformance of BPR-US algorithm and PFU-CFD algorithm. The results show thatARTD can recovery the traffic data effectively; BPR-US algorithm can reach a pointwhere traffic load is not so high and most users can satisfy the guidance results;PFU-CFD algorithm can guarantee the fairness of the system in multi-time guidance.