A trustworthiness-based vehicular recruitment scheme for information collections in Distributed Networked Systems

Because of high mobility, large number of vehicles are utilized to achieve timely and quality-based information in the smart Internet of Things, which has formulated into a dynamic Distributed Networked Systems (DNS). However, designing a vehicular recruitment scheme to enhance a security-based DNS is challenging since it is hard to judge trustworthiness values of vehicular sensors. Therefore, in this paper, a novel vehicular trust evaluation scheme is proposed to analyze and supervise the data collected by vehicular sensors with a trust and low-cost style. To obtain vehicular trusts, the proposed scheme that considers time factor and gap between sensed data and real data is designed to calculate trustworthiness of vehicles. Moreover, sensing data in the vehicle sparse regions has more contributions because of its rareness. Thus, to inspire vehicles to sense data in the vehicle sparse regions, a trustworthiness-based gradient pricing method is designed to pay rewards for the vehicular sensors. Finally, with real vehicular GPS datasets, simulation results demonstrate that the proposed scheme can improve accuracy rate of data sensing by 37.72% and can improve data quality by 76.95%. By incentive pricing method, coverage ratio of data sensing is improved by 13.1%. In general, performances of the proposed scheme can be improved by 19. 39% to 22.32% approximately. Future works focus on improving information security by advanced machine learning methods in the dynamic DNS.