Collaborative data collection with hybrid vehicular crowd sensing in smart cities

Vehicular crowd sensing (VCS) is an emerging data collection paradigm in smart cities, where vehicles are incentivized to perform complex urban sensing tasks. Original VCS is difficult in supporting large-scale and fine-grained urban data collection. In this paper, we propose a hybrid VCS paradigm, where VCS network and wireless sensor network (WSN) cooperate to provide urban data collection service and gain rewards responding. The cooperation of VCS network and WSN can significantly expand the sensing range and improve the sensing quality as well. In this hybrid paradigm, data center, VCS network, and WSN all aim to maximize their revenues, individually. We utilize the methodology of Stackelberg game to design an optimal collaborative strategy based on a three-party model, where both VCS network and WSN act as the leaders while the data center acts as the follower. We theoretically prove that the three-party game can converge to a unique Nash equilibrium. At the Nash equilibrium, three players choose their best response to obtain their maximum revenues, respectively. The simulation results validate the effectiveness of the optimal collaborative strategy.