A Trajectory Privacy-Preserving Scheme Based on Transition Matrix and Caching for IIoT

Shiwen Zhang,Biao Hu,Wei Liang,Kuan-Ching Li,Al-Sakib Khan Pathan

DOI: https://doi.org/10.1109/jiot.2023.3308073

IF: 10.6

2023-01-01

IEEE Internet of Things Journal

Abstract:With the increasing integration of Location-Based Services (LBS) into various societal domains, location privacy preservation has emerged as a pivotal concern. In most continuous LBS privacy protection approaches, the user needs to send a query to an untrusted Location Service Provider (LSP) to request the corresponding query results, and these results are discarded immediately after being used. This leads to similar queries in the future having to be sent to the LSP again, which increases the risk of privacy leakage when facing the LSP. To solve these issues, caching techniques are typically used to provide answers to users’ future queries. However, minimizing the interaction with the LSP is a challenge. Here, we propose a trajectory privacy-preserving scheme based on a transition matrix and caching (TMC) scheme for continuous LBS in the Industrial Internet of Things (IIoT). It employs multi-level caching to reduce the risk of exposing sensitive information to untrusted entities. We designed a transition matrix to predict the user’s next query location and simplify the computation complexity. We designed a cloaking set generation algorithm by considering transition entropy, location prediction, data freshness, and cache contribution degree to enhance user location privacy and improve the cache hit rate. The security analysis demonstrates how the TMC scheme resists attacks from both internal and external entities and ensures robustness in trajectory privacy. The experimental results show that the proposed TMC scheme can provide a higher level of privacy protection and lower system overhead compared to several previous schemes.

computer science, information systems,telecommunications,engineering, electrical & electronic