.png){kind=icon}
.png){kind=icon}
Return
Activity-based capability updating method for task assignment in mobile crowdsensing
.png){kind=icon}
PRE
AI
DOI：10.1016/j.comnet.2025.111304.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Task allocation is a critical process for mobile crowdsensing. In this process, the platform assigns tasks uploaded by requesters to suitable workers based on specific criteria. To ensure maximum benefit, crowdsensing platforms must select enough workers to complete these tasks with high quality. Therefore, it is essential for the platform's profitability to select reliable and punctual workers. Current approaches rely on intrinsic attributes for winner selection, which can lead to issues with unstable perceived quality. In this study, we propose a worker capability update equation that takes into account workers' temporal and spatial activity levels to select stable and high-quality contributors. Additionally, we introduce a metric based on normal distribution to evaluate workers' task completion rates. By utilizing historical task performance, we can select workers for each task, mitigating the impact of quality fluctuations on the platform's profits. We compared our method with four baseline methods using real datasets, and all results demonstrated its efficiency.
Keywords:
Mobile crowdsensing
Opportunistic task assignment
Profit maximization
Journal
IF:
4.6
Papers:
1.5K
Citations:
1.6W
Organization
No organization information available