Geo-indistinguishable Mechanisms for Spatial Crowdsourcing via Multi-Objective Evolutionary Optimization

Industrial Internet of Things (IIoT) has exploded key revolutions in several leading industries, such as energy, agriculture, mining, transportation, and healthcare. Due to the nature of high capacity and fast transmission speed, 5G plays a pivot role in enhancing the industrial procedures, practices and guidelines, such as crowdsourcing, cloud outsourcing and platform subcontracting. Spatial crowdsourcing (SC)-servers (such as offered by DiDi, MeiTuan and Uber) assign different tasks based on workers' location information.However, SC-servers are often untrustworthy and have the threat of revealing workers' privacy. In this paper, we introduce a framework Geo-MOEA (Multi-Objective Evolutionary Algorithm) to protect location privacy of workers involved on SC platform in 5G environment. We propose an adaptive regionalized obfuscation mechanism with inference error bounds based on geo-indistinguishability (a strong notion of differential privacy), which is suitable for the context of large-scale location data and task allocations. This offers locally generated pseudo-locations of workers to be reported instead of their actual locations.Further, to optimize the trade-off between SC service availability and privacy protection, we utilize MOEA to improve the global applicability of the mechanism in 5G environment. Finally, by simulating the location scenario, the visual results on experiments show that the mechanism can not only protect location privacy, but also achieve high availability of services as desired.