利用覆盖范围和声誉的移动众包感知联合约束激励机制算法。

College of Computer Science and Technology, Chang Chun University of Science and Technology, Changchun 130022, China.

Petrochina Research Institute of Petroleum Exploration and Development, Beijing 100083, China.

Sensors (Basel). 2020 Aug 11;20(16):4478. doi: 10.3390/s20164478.

Selection of the optimal users to maximize the quality of the collected sensing data within a certain budget range is a crucial issue that affects the effectiveness of mobile crowdsensing (MCS). The coverage of mobile users (MUs) in a target area is relevant to the accuracy of sensing data. Furthermore, the historical reputation of MUs can reflect their previous behavior. Therefore, this study proposes a coverage and reputation joint constraint incentive mechanism algorithm (CRJC-IMA) based on Stackelberg game theory for MCS. First, the location information and the historical reputation of mobile users are used to select the optimal users, and the information quality requirement will be satisfied consequently. Second, a two-stage Stackelberg game is applied to analyze the sensing level of the mobile users and obtain the optimal incentive mechanism of the server center (SC). The existence of the Nash equilibrium is analyzed and verified on the basis of the optimal response strategy of mobile users. In addition, mobile users will adjust the priority of the tasks in time series to enable the total utility of all their tasks to reach a maximum. Finally, the EM algorithm is used to evaluate the data quality of the task, and the historical reputation of each user will be updated accordingly. Simulation experiments show that the coverage of the CRJC-IMA is higher than that of the CTSIA. The utility of mobile users and SC is higher than that in STD algorithms. Furthermore, the utility of mobile users with the adjusted task priority is greater than that without a priority order.

在一定预算范围内选择能够最大限度地提高所收集传感数据质量的最优用户，是影响移动众包感知（MCS）有效性的关键问题。目标区域内移动用户（MUs）的覆盖范围与传感数据的准确性有关。此外，MUs 的历史声誉可以反映他们之前的行为。因此，本研究提出了一种基于 Stackelberg 博弈论的覆盖和声誉联合约束激励机制算法（CRJC-IMA），用于 MCS。首先，利用移动用户的位置信息和历史声誉选择最优用户，从而满足信息质量要求。其次，应用两阶段 Stackelberg 博弈分析移动用户的传感水平，并获得服务器中心（SC）的最优激励机制。在移动用户最优响应策略的基础上，分析并验证纳什均衡的存在性。此外，移动用户将及时调整任务的优先级，使所有任务的总效用达到最大值。最后，使用 EM 算法评估任务的数据质量，并相应地更新每个用户的历史声誉。仿真实验表明，CRJC-IMA 的覆盖范围高于 CTSIA。移动用户和 SC 的效用高于 STD 算法。此外，调整任务优先级的移动用户的效用大于没有优先级顺序的移动用户的效用。