A blockchain-based creditable and distributed incentive mechanism for participant mobile crowdsensing in edge computing.

Abstract

With the popularization of portable smart devices, the advance in ubiquitous connectivity and the Internet of Things (IoT), mobile crowdsensing is becoming one of the promising applications to acquire information in the physical world of edge computing and is widely used in Smart Cities. However, most of the existing mobile crowdsensing models are based on centralized platforms, which have some problems in reality. Data storage is overly dependent on third-party platforms leading to single-point failures. Besides, trust issues seriously affect users' willingness to participate and the credibility of data. To solve these two problems, a creditable and distributed incentive mechanism based on Hyperledger Fabric (HF-CDIM) is proposed in this paper. Specifically, the HF-CDIM combines auction, reputation and data detection methods. First, we develop a multi-attribute auction algorithm with a reputation on blockchain by designing a smart contract, which achieves a distributed incentive mechanism for participants. Second, we propose a K-nearest neighbor outlier detection algorithm based on geographic location and similarity to quantify the credibility of the data. It is also used to update the user's reputation index. This guarantees the credibility of sensing data. Finally, the simulation results using real-world data set verify the effectiveness and feasibility of above mechanism.