Research On Routing Algorithm Based On Vehicle Social Characteristics In Vehicle Social Network

The concept of ITS (Intelligent Transportation System) does not represent only a unrealistic vision, but a viable reality due to a new type of emerging wireless ad hoc network for vehicular environment i.e.,VANET (Vehicular Ad Hoc Network). VANET is a special kind of MANET (Mobile Ad Hoc Network), characterized by high (variable)vehicle speed, hostile propagation environment, and quickly changing network topologies. Due to the inseparable relationship between a vehicle and its user, VANET also has some social characteristics found in social network. Social characteristics and human behaviors largely impact on VANET, and this arises to the concept of VSN (Vehicle Social Network).VSN is the combination of social network and vehicular network, the social properties of the mobile vehicles can be exploited to design efficient routing algorithms in Vehicle Social Network.Firstly, this paper surveys the emerging applications, important challenging issues and existing routing algorithms of VSN. Secondly,according the advantages and shortcomings of existing routing algorithms,it presents a new routing algorithm called IntAware (An Interest-based community aware routing algorithm). In IntAware, nodes with same interests are considered into the same community. IntAware uses different routing strategies in intracommunity routing and intercommunity routing.For intracommunity routing, the social tie metric convoluting social centrality and similarity is used to decide the next forwarding node,which can solve the problems of dead-end and blind-spot commonly appeared in the utility-based routing algorithm. For intercommunity routing, the community density metric is used to decide the forwarding node. The efficiency of IntAware is evaluated by the ONE, Evaluation results show that IntAware obtains better routing performance than Epidemic, Prophet and BeeInfo in terms of overhead, delivery ratio and average hop-count, although its average latency is not the best, anyway this is acceptable in many applications.Additionally, we analyze the vehicle trace of public transportation,SUVnet-trace and propose GeoSmar, a social-and mobile-aware geocast routing algorithm for urban VANET from the perspectives of delay tolerant network and social network. GeoSmar exploits two levels of vehicle’s (macro and micro) social mobility characteristics and the contact information to route the message to the target geographical region,Simulation experiment on GeoSmar uses the SUVnet-trace dataset to narrow the gap between real trace and mobility model. The results verified the performance supremacy of GeoSmar.