On Coalitional and Non-Coalitional Games in the Design of User Incentives for Dependable Mobile Crowdsensing Services

In the Era of massive connectivity, crowdsensed data services are becoming attractive so to acquire sensory data from smart mobile devices that are recruited for cooperative collection of sensed data through embedded sensors. Despite its benefits to offer sensing as a service by taking advantage of the non-dedicated nature MCS systems, it remains certain challenges such as user privacy and data trustworthiness, which could affect the intention of an individual to participate in the sensing phenomena. Furthermore, in the presence of physical tampering or falsification of sensor readings, trustworthiness of crowd-sensed data arises as a big concern. Game theoretic solutions are popular to address the utility maximization needs of participants, and the trustworthiness objectives of the MCS platforms. We revisit coalitional game formation and subgame perfect equilibrium-based concepts to motivate more users for truthful participation, and present a comparative study of these game models from the standpoint of platform utility, user utility and data trustworthiness. Our case studies show that the subgame perfect equilibrium-based model can provide higher user utility and up to 30% higher platform utility in comparison to the coalitional recruitment model in the presence of a large adversary population in an MCS campaign. Furthermore, regardless of the adversary population, repeated subgame-based model results in lower disinformation probability in the system when compared to coalitional game-based user incentivization.