A vehicle clustering algorithm for information propagation by inter-vehicle communications

There are some cases regarding information propagation over a wide area, such as Vehicle-to-Vehicle (V2V) communications, in which an end-to-end path does not exist, due to the non-uniform density of vehicles. In order to solve this problem, Delay Tolerant Network (DTN) architectures have been studied to facilitate message delivery in V2V communications. However, existing message delivery methods for the purpose of information sharing do not consider the efficiency of message passing communications. As a result, information can be lost, due to a decrease in the efficiency of information propagation. In order to increase the efficiency of information propagation, we propose a vehicle clustering algorithm, based on the similarity of geographic locations and trajectories. This contributes to efficient message passing communication. In order to evaluate our proposal, we carry out a simulation. The simulation results confirm that our proposed method provides higher message reachability, and utilizes bandwidth more efficiency than existing methods.