Jiangsu Key Laboratory of Big Data Security and Intelligent Processing, Nanjing University of Posts and Telecommunications, Nanjing 210023, China.
Department of Computer Science, Colorado School of Mines, Golden, CO 80401, USA.
Sensors (Basel). 2020 Feb 2;20(3):805. doi: 10.3390/s20030805.
The recent development of human-carried mobile devices has promoted the great development of mobile crowdsensing systems. Most existing mobile crowdsensing systems depend on the crowdsensing service of the deep cloud. With the increasing scale and complexity, there is a tendency to enhance mobile crowdsensing with the edge computing paradigm to reduce latency and computational complexity, and improve the expandability and security. In this paper, we propose an integrated solution to stimulate the strategic users to contribute more for truth discovery in the edge-assisted mobile crowdsensing. We design an incentive mechanism consisting of truth discovery stage and budget feasible reverse auction stage. In truth discovery stage, we estimate the truth for each task in both deep cloud and edge cloud. In budget feasible reverse auction stage, we design a greedy algorithm to select the winners to maximize the quality function under the budget constraint. Through extensive simulations, we demonstrate that the proposed mechanism is computationally efficient, individually rational, truthful, budget feasible and constant approximate. Moreover, the proposed mechanism shows great superiority in terms of estimation precision and expandability.
近年来，人们携带的移动设备得到了极大的发展，这推动了移动众包感知系统的飞速发展。大多数现有的移动众包感知系统都依赖于深度云的众包服务。随着规模和复杂度的不断增加，移动众包感知系统有向边缘计算范例增强的趋势，以降低延迟和计算复杂度，并提高可扩展性和安全性。在本文中，我们提出了一种综合解决方案，以激励战略用户在边缘辅助的移动众包感知中为真相发现做出更多贡献。我们设计了一个激励机制，包括真相发现阶段和预算可行的逆向拍卖阶段。在真相发现阶段，我们在深度云和边缘云中对每个任务进行真实估计。在预算可行的逆向拍卖阶段，我们设计了一个贪婪算法，在预算约束下选择最大化质量函数的获胜者。通过广泛的仿真，我们证明了所提出的机制在计算效率、个体理性、真实性、预算可行性和常数近似方面具有优势。此外，所提出的机制在估计精度和可扩展性方面表现出了巨大的优势。
