Decentralized Dynamic Participation in Participatory Sensing: A Correlated-Equilibrium Game Approach

Abstract

Participatory sensing has become a compelling and viable paradigm for large scale sensing data collection in recent years, with smartphone workers recruited from a public crowd. Since the quality of the sensing data is directly related to the performance of the workers, it is then crucial to enable effective participation for a large number of workers. In this paper, we propose a decentralized dynamic participation game framework for participatory sensing systems with heterogeneous sensing processes and smartphone workers. Based on economic studies, we characterize the behaviour of workers with the idea of correlated equilibrium. We propose a regret matching based participation algorithm to track the set of correlated equilibria in a distributed manner, where the knowledge of competitors' preference information is not required. Simulation results demonstrate that our algorithm achieves good convergence and effective participation.