Remote Driver Identification Using Minimal Sensory Data

This letter proposes a driver identification algorithm developed for using minimal information transmitted in vehicle-to-vehicle (V2V) environment based on the assumption that each driver has his/her own unique driving behavior characteristics hidden in conventional driving behaviour signals. Performance evaluation using real-world data shows that the proposed algorithm accurately identifies three drivers in 15 trips using only two acceleration signals at a sampling rate much lower than the one used in previous study. Furthermore, it is shown that the proposed algorithm is able to perform driver identification within a small connectivity time in a scenario where driving behavior signals of are transmitted via a V2V communication environment in a streaming manner.