A mobile crowdsensing (MCS) platform motivates employing participants from the crowd to complete sensing tasks. A crucial problem is to maximize the profit of the platform, i.e., the charge of a sensing task minus the payments to participants that execute the task. In this article, we improve the profit via the data reconstruction method, which brings new challenges, because it is hard to predict the reconstruction quality due to the dynamic features and mobility of participants. In particular, two Profit-driven Online Participant Selection (POPS) problems under different situations are studied in our work: (1) for S-POPS, the sensing cost of the different parts within the target area is the Same. Two mechanisms are designed to tackle this problem, including the ProSC and ProSC+. An exponential-based quality estimation method and a repetitive cross-validation algorithm are combined in the former mechanism, and the spatial distribution of selected participants are further discussed in the latter mechanism; (2) for V-POPS, the sensing cost of different parts within the target area is Various, which makes it the NP-hard problem. A heuristic mechanism called ProSCx is proposed to solve this problem, where the searching space is narrowed and both the participant quantity and distribution are optimized in each slot. Finally, we conduct comprehensive evaluations based on the real-world datasets. The experimental results demonstrate that our proposed mechanisms are more effective and efficient than baselines, selecting the participants with a larger profit for the platform.

中文翻译：

---

压缩移动众测中在线参与者选择过程中的利润优化

移动人群感知 (MCS) 平台激励从人群中雇佣参与者来完成感知任务。一个关键问题是最大化平台的利润，即传感任务的费用减去执行任务的参与者的付款。在本文中，我们通过数据重建方法提高了收益，这带来了新的挑战，因为由于参与者的动态特征和流动性，很难预测重建质量。特别是，在我们的工作中研究了两个不同情况下的利润驱动的在线参与者选择（POPS）问题：（1）对于S-POPS，目标区域内不同部分的感知成本是相同的。设计了两种机制来解决这个问题，包括 ProSC 和 ProSC+。前者机制结合了基于指数的质量估计方法和重复交叉验证算法，后者机制进一步讨论了选定参与者的空间分布；(2) 对于V-POPS，目标区域内不同部分的感知成本是多种多样的，这使其成为NP-hard问题。提出了一种称为 ProSCx 的启发式机制来解决这个问题，其中搜索空间变窄，并且每个时隙中的参与者数量和分布都得到了优化。最后，我们基于真实世界的数据集进行综合评估。实验结果表明，我们提出的机制比基线更有效和高效，为平台选择了利润更大的参与者。