Crowdsensing-Based Consensus Incident Report for Road Traffic Acquisition
Crowdsensing-Based Consensus Incident Report for Road Traffic Acquisition
复制标题
基于群体感知的道路交通采集共识事件报告
DOI：
10.1109/tits.2017.2750169
复制
发表时间：
2018-08-01
影响因子：
8.5
通讯作者：
Wang, Xinbing
中科院分区：
文献类型：
--
作者：
Wang, Xiong;Zhang, Jinbei;Wang, Xinbing
实时的道路交通信息给驾驶员带来了极大的便利。最近的移动群智感知范式使得各种道路信息获取成为可能。但信息的准确性无法保证，也缺乏适当的激励机制。在本文中，我们研究了根据大量未知贡献者的报告提取实际道路交通信息的问题。为了高概率地获得准确的道路交通结果，我们建立了一个声誉系统来评估每个贡献者的可靠性，该系统考虑了位置和时间偏差因素。我们还设计了激励机制，以促使每个合格的贡献者如实报告。此外，我们改进了现有的答案推理方法，并以有效的方式得出正确的结果。进行了广泛的模拟来评估所提出的算法。
Real-time road traffic information brings great convenience for drivers. Various road information acquisitions are enabled by recent mobile crowdsensing paradigm. However, the accuracy of information can not be guaranteed, and appropriate incentive mechanism is still unavailable. In this paper, we study the problem of extracting the actual road traffic information according to the reports from an amount of unknown contributors. To obtain the accurate road traffic result with high probability, we establish a reputation system to evaluate the reliability of each contributor, which takes both location and time deviation factors into account. We also design an incentive mechanism to elicit the truthful report of each qualified contributor. Furthermore, we improve the existing answer inference methods and derive the correct result in an efficient way. Extensive simulations are carried out to evaluate the proposed algorithms.