MAIM: A Novel Incentive Mechanism Based on Multi-Attribute User Selection in Mobile Crowdsensing

In the user selection phase of mobile crowdsensing, most existing incentive mechanisms focus on either single-attribute selection or random selection, which possibly lead to serious consequences such as low user enthusiasm, decreased task completion rate, and increased cost of platform consumption. To tackle these issues, in this paper, we propose a novel incentive mechanism MAIM, which is based on multi-attribute user selection and participation intention analysis function in mobile crowdsensing. In this mechanism, the sensing platform employs the analytic hierarchy process to determine the weights of three attributes: participation threshold, cost, and reputation. The weight calculation results of each sensing user with respect to each attribute are then integrated to obtain the sorted weight of each user, with which the sensing platform will then obtain the optimal user set. From the users' perspective, they can autonomously decide whether to accept task processing requests, as enabled by the participation intention analysis function, thereby voiding the absolute authority and control of the sensing platform over users and achieving a two-way selection between the sensing platform and the sensing users. Furthermore, the sensing platform establishes a score-based reputation reward to inspire active performers and utilizes a punishment mechanism to overawe malicious vandals, which substantially helps activize enthusiasm of user participation and improve sensing data quality. Simulation results indicate that the proposed MAIM has significantly improved the sensing task completion ratio and the budget surplus ratio compared with the existing incentive mechanisms in mobile crowdsensing.