Incentivizing fairness-aware task allocation in mobile crowdsensing with sweep coverage and stability control

Abstract

Mobile crowdsensing (MCS) harnesses the sensing capabilities of sensors built into a large number of smart devices to collect and analyze sensory data, which can be used by a large number of mobile participants to perform numerous sensing tasks. Sweep coverage and stability control are two key issues that need to be solved in task allocation in MCS, because improper matching of tasks and participants, as well as task overload or idleness will slump social welfare. However, most existing work made an optimistic assumption that all participants unconditionally participate in MCS, without considering the selfishness and rationality of participants in practical scenarios. To tackle these issues, this paper proposes a fairness-aware task allocation policy with sweep coverage and stability control, which consists of an online rating protocol and a stability control scheme. For the first module, we integrate the quality of sensing, rating update, collusion identification, and payment determination to develop an online rating protocol to deal with the “free-riding” and collusion of participants simultaneously. For the second module, despite the unpredictable future information of sensing tasks and participants, we design a stability control scheme that only relies on currently information to make the online control independent strategies for maximizing social welfare and balancing network stability in a proportional fairness, which can maintain system stability and achieve a time average social welfare within O(1∕V) that is arbitrarily close to the optimum for any tunable parameter V>0. Finally, through rigorous theoretical analysis and experimental comparison with two benchmarks, the correctness and efficiency of our proposed policy is jointly demonstrated.