An incentive mechanism design for mobile crowdsensing with demand uncertainties

Abstract

Mobile crowdsensing (MCS) has shown great potential in addressing large-scale data sensing problem by allocating sensing tasks to pervasive mobile users (MU). The MUs will participate in the MCS if they can receive sufficient compensation. Existing work has designed lots of incentive mechanisms for MCS, but ignores the MUs’ resource demand uncertainties that is critical for resource-constrained mobile devices. In this paper, we propose to design an incentive mechanism for MCS by taking the MUs’ own resource demand into the economic model. As different MUs will have different behavior, they will participate in the MCS with different levels. Based on this idea, we formulate the incentive mechanism by using the Stackelberg game theory. Furthermore, a dynamic incentive mechanism (DIM) based on deep reinforcement learning (DRL) approach is investigated without knowing the private information of the MUs. It enables the SP to learn the optimal pricing strategy directly from game experience. Finally, numerical simulations are implemented to evaluate the performance and theoretical properties of the proposed mechanism and approach.