SenseNow: A Time-Dependent Incentive Approach for Vehicular Crowdsensing

Abstract

This paper presents an incentive mechanism for vehicular crowdsensing (VCS). Here, a platform selects a set of spots or Places of sensing Interest (PsI) and outsources the collection of data from these places. In particular, the platform is interested in collecting data from most of the PsIs (spatial coverage) at regular and well-spread time intervals (temporal coverage). Although spatial coverage is a natural by-product of this approach, our main focus is to reach temporal coverage. To this goal, we model the interaction between participants (vehicles) as a non-cooperative game in which vehicles are the players, and the time to sample at a given PsI is the players’ strategy. Here, vehicles are rewarded for deviating from their pre-planned paths and visiting a set of PsIs. The rewarding formula is designed such that selfish vehicles trying to maximize their reward will collect high temporal coverage data. In particular, this paper analyses the effects of increasing the number of vehicle deviations on the utilities of both vehicles and the platform.