Vehicle-to-Vehicle Content Distribution In Two-way Vehicular Network Based On Network Coding

Internet of Vehicles,which supports ubiquitous information exchange and content sharing among vehicles,is the key to the realization of intelligent transportation.Content distribution,such as route planning,collision warning,online video and traffic monitoring,is a very important application in the network of vehicles.This paper mainly studies the distribution of content combined with network coding and cache forwarding in a two-way vehicular network.This paper considers a relatively simple situation first,that is,there are two different data content disseminating on two ways.Using the cache forwarding capability of mobile vehicles,this paper proposes a mechanism that combines network coding with simple storage and forwarding to increase the content disseminating rate and evaluate the rate gain under different conditions through simulation.Under this mechanism,when the vehicle node passes through the disseminating node in the opposite way,the received network coded data packets are stored.When it becomes the dissemiting node,this node forwards packets that vehicles on opposite roadway are interested in and transmits packets vehicles on same roadway are interested in at same time using directional antennas.Then this paper considers a more general case,that is,the distribution of multiple different contents in a two-way network.This paper proposes a pre-cached two-way multiple content distribution scheduling mechanism based on network coding and compares the performance of equal caching and popularity-based caching.All vehicle nodes pre-cache a part of the network coding packets.When a node needs one of these content,it first obtains the coding packets from the nodes of the same direction in the single-hop communication range,and then obtains coding packets from the opposite direction way.If the vehicle node does not decode the content in V2V(vehicle-to-vehicle)phase,it finally downloads the remaining required data packets from the base station or the roadside unit.Simulation results show that the popularity-based caching strategy outperforms the equal cache when the vehicle is traveling at high speed.