A Binary Rating Protocol For Crowdsensing Applications

Crowdsensing, as a novel method of data collecting, environment sensing and information providing, has been widely used in place of the traditional wired sensor networks and wireless sensor network to tackle real-world challenges by leveraging mobile devices. Unfortunately, the practicality of crowdsensing can be tampered with antisocial behavior, such as free-riding, false-reporting and whitewashing. Hence, incentive mechanisms are needed to compel self-interested users to contribute well-behavior. In this paper, we propose a rating protocol by integrating the reputation and pricing schemes based on game theory for crowdsensing, in which each participant is labeled with a binary rating to represent its social status, and users are encouraged to contribute good behaviors to increase their ratings and indirectly receive higher rewards. By quantifying the sufficient and necessary conditions under which users comply with the rating protocol in their self-interests to do so, we formulate the rating protocol design problem, and further verify the protocol sustainability. We rigorously analysis the optimal social welfare, as well as the optimal design parameters. The proposed protocol design for crowdsensing applications can achieve social benefit optimization and prevent the whitewashing problem.