SecureLoc: A fully homomorphic encryption-based privacy protection scheme for location-based services

Abstract

As Internet of Things (IoT) technology advances, a growing number of devices can access real-time location information and engage with other devices and platforms. Consequently, this expansion enriches the data sources and application scenarios for Location-Based Services (LBS). The computational tasks of LBS are often outsourced to a third-party service (TPS) for processing in order to improve computational efficiency on users’ devices. However, sensitive and private data stored with a semi-honest TPS poses the risk of data abuse or data leakage. In this paper, we propose a robust privacy-preserving scheme called SecureLoc within outsourced computing environments. Utilizing the collaborative capabilities of the TPS and the Trajectory Matching Server (TMS), we present a fully homomorphic encryption approach to protect the privacy of location and sensitive information. Specifically, we present an improved CKKS-based trajectory comparison algorithm that ensures trajectory matching without exposing sensitive plaintext data. In addition, by utilizing complex numbers to store location coordinates and ciphertext expansion, we greatly improve the computational efficiency. We also combine the K-anonymity algorithm with CKKS to further enhance the protection of user privacy by anonymizing and generalizing sensitive information such as phone numbers, ID numbers, and LBS request times. Finally, we prove SecureLoc is secure against semi-honest TPS and malicious eavesdroppers, and demonstrate that our method outperforms other state-of-the-art methods in terms of security, feasibility, and accuracy.