Incentive mechanisms for crowdsensing: safeguarding against malicious behaviors

Abstract The high efficiency of mobile crowdsensing (MCS) relies heavily on motivating users to participate in sensing tasks. Designing an auction-based incentive mechanism is a widely adopted approach. However, platforms operating in the unrestricted Internet environment are inevitably vulnerable to various types of malicious behaviors. While most existing studies focus solely on countering a single type of malicious behaviors, their approaches often lead to a decline in task acceptance rates, ultimately impacting the system’s utility. To address this challenge, we propose an incentive mechanism to resist multiple malicious user behaviors in a reverse auction, including monopoly and malicious competition. The PT-IM model is first introduced to identify and exclude monopolistic users through the calculation of tolerance price and user capability. Additionally, a novel task area division method is implemented within PT-IM to improve the task acceptance rate. Building on this foundation, we develop the enhanced model EPT-IM to further mitigate malicious competition among users through primary selection and secondary selection conditions. We conduct both theoretical and experimental analysis to evaluate EPT-IM. The results demonstrate that the proposed mechanism effectively resists malicious behaviors of users and surpasses other incentive mechanisms in terms of overall performance.