Sensing Vehicle Selection Scheme Optimization in Vehicular Crowdsensing

Vehicular crowdsensing (VC) is one of the applications of mobile crowdsensing technology. Vehicular crowdsensing can be an effective method for urban traffic sensing applications by collecting data in urban road networks through ubiquitous sensor-mounted vehicles. However, due to the limited network resources and the randomness of automobiles, the quality of service (QoS) of VC cannot be effectively guaranteed. In this paper, a novel efficient model is proposed to evaluate the QoS of VC, which considering the sensing coverage and road-network topology (the importance of sensed information is related to road-network topology). In addition, we propose a novel node selection method, and evaluate the performance of it through comparing with other exiting methods. The results demonstrate that the proposed model is effective in better evaluating the QoS of VC, and the novel node selection method is more useful in improving the QoS of VC than other methods.