Incentive Mechanism Design Based on Stochastic Game for Multi-modality Crowd Sensing

Mobile Crowd Sensing (MCS) has emerged as a paradigm for data collection and analysis due to the pervasive smart devices and the need for sensing data. The consumption of power, network flow and privacy exposure limits users' participation. A large number of incentive mechanisms are needed. However, the existing mechanisms do not consider the uncertainty and the probability of users' behaviors. To address this issue, we develop an incentive mechanism based on stochastic game with the uncertainty of the users' behaviors in this paper. We try to incentive the participant level of the task participants and design strategies for the participants to select suitable tasks to perform. We investigate the actions of task providers and task participants, and solve the probabilities of behaviors while guaranteeing the minimum payoff of the participants. The numerical results show that our approach is 19% better than the existing method Frugal-OMZ on the payoff and 44.5% on the number of the tasks.