Dynamic Incentive for Reliable MCS Participant Selection
Xinqiang Ma,Yi Huang,Qiang Li,Biao Huang,Xiaoye Liu,Jia Liu
DOI: https://doi.org/10.1109/jiot.2023.3266022
IF: 10.6
2023-01-01
IEEE Internet of Things Journal
Abstract:Mobile Crowdsensing (MCS), which relies on a large number of ordinary participants to complete sensing tasks, has become a new data collection paradigm in the Internet of Things (IoT). In the MCS system, how to select reliable participants to participate in the sensing task and provide appropriate rewards for participants is one of the main problems faced by MCS. This paper proposes a reliable participant selection strategy driven by dynamic incentives. The whole participant selection process is divided into multiple stages. First, the platform calculates participants reliability by considering task completion rate and task pass rate. Then, when the platform releases tasks at each stage, it considers the task deadline and the task completion progress to quantify the task requirements, and then determines the maximum price of the tasks at that stage. Finally, the platform selects participants who meet the requirements for different tasks at each stage, and conducts bargaining games with participants to determine appropriate rewards. Considering the situation that participants quit midway and the task cannot be completed, the platform adopts a task delegation mechanism to ensure the effective completion of the task. The results show that the proposed strategy can guarantee the task completion rate and data quality, and optimize the reward of participants.
computer science, information systems,telecommunications,engineering, electrical & electronic