Cache Strategies In VANETs Via Named Data Networking

With the rapid development of intelligent transportation and the Internet of Things, Vehicular Ad Hoc Networks (VANETs) is more and more attractive. The VANETs have advanced the development of modern transportation by realizing the communication of vehicles to vehicles,vehicles to the infrastructure and vehicles to the Internet.The IP-based vehicle network, due to the high mobility of the vehicle nodes, rapid changes in the topology, will result in intermittent link and easy break. Named Data Networking (NDN), which uses named data instead of IP addresses, does not care about the host location. The nodes in NDN have capabilities of caching and characteristics of multi-source data, which is more suitable for requirements of VANETs. But in the existing NDN cache decision and replacement strategies, the scenario is too single, not fully consider the mobility of the vehicle nodes and wireless radio features and not fully show advantages of the network cache, such as fast retrieval and efficient access to data.In this thesis, the Naming Data Network architecture is merged in VANETs. According to the requirement of the effectiveness and high efficiency of the communication in VANETs, this thesis designs the cache strategy of NDN in the vehicle networking environment.In the aspect of cache decision, this thesis designs a distributed cache decision strategy DPC based on the preference of users, the importance of nodes and relative motion. In the simulation, the vehicle motion model simulator SUMO (Simulation of Urban Mobility) and the discrete event simulator NS-3 are combined to simulate the cache performance of the algorithm in the highway scene. The results of experiments indicate that the proposed DPC algorithm has higher cache hit ratio and diversity of content, lower latency than Always cache strategy and fixed probability cache strategy.In terms of cache replacement strategy, this thesis proposes a cache replacement strategy VCRS based on the utility function, considering the contradiction between limited cache capacity and infinite cache content.Under the same conditions, compared with the lru,lfu and fifo, the proposed VCRS algorithm improves the cache hit ratio , reduces the delay and the average number of hop .