Double Cloak Area Approach for Preserving Privacy and Reliability of Crowdsourcing Data

Abstract

[EN] Crowdsourcing has emerged as a pivotal data source for diverse smart city applications, ranging from health and traffic to security and safety. However, the integration of users' location data in crowdsourced information poses a significant privacy challenge. Current privacy protection approaches of location-based services have become inadequate to face the evolving attackers' techniques and tools. Moreover, these protection methods ignored the issue of preserving the accuracy and reliability of data. This paper introduces a novel approach, termed Double Cloak Area (DCL-Ar), designed to effectively safeguard users' location privacy and ensure the reliability of data based on crowdsourcing. DCL-Ar differentiates by offering dual-layer protection for identity. The first layer involves users creating an initial cloak zone, while the second layer utilizes fog nodes to establish an extended cloak zone. Furthermore, the proposed method introduces three distinct scenarios for managing collaboration among fog nodes to select the optimal anonymizer and address the limitations of existing protection methods which are related to saving the reliability and the accuracy of data. DCL-Ar maintains maximum entropy, achieving complete uncertainty about user locations, thereby ensuring a high level of privacy protection. Through simulation and comparative analysis, the efficacy of the proposed approach is demonstrated where it provides a superior privacy level without significant performance. Experimental results demonstrate that DCL-Ar outperforms traditional methods, improving cache hit ratios and response times while reducing server query loads. Specifically, our approach reduces the number of queries sent to the service provider (SP) by up to 50% compared to existing methods and maintains a high cache hit ratio of nearly 100% over time. It further impacts on the traditional cloak-area and other protection approaches.