A data quality index based incentive mechanism for smartphone crowdsensing
Tiancheng Hu, Tao Yang, Bo Hu · 2016
Crowdsensing is an emerging paradigm which recruits smartphone users to contribute their sensor resources for data collecting tasks. To achieve better service performance, a proper incentive mechanism should be designed to stimulate user participation. Though there have been plenty of works on incentive mechanism design in literature, most of them ignore the submitted data quality, which actually has significant influence on the crowdsensing performance. In this paper, the metric of data quality is introduced to measure the value of user to the tasks, based on which a social welfare maximization problem is formulated and a novel incentive mechanism through reverse auction is proposed. Specifically, The crowdsourcer considers user bidding profiles and their sensing quality together, and the quality requirements of tasks are used as constraints to ensure better sensing performance. Through rigorous theoretical analysis, the proposed mechanism is proved to be computational efficient, individual rational and truthful. Meanwhile, the approximate optimal solution within a constant factor is achieved. Simulation results verified the effectiveness of the proposed mechanism.