The issue of user privacy in the context of collaborative positioning is addressed in this work, wherein information is passed between and processed by multiple cooperative agents, with the goal of achieving high levels of positioning accuracy. In particular, this paper discusses three privacy-preserving schemes in the context of differential global navigation satellite system (GNSS)-based and GNSS-based cooperative positioning methods. The discussed architectures provide the same positioning results, while yielding different levels of privacy to the cooperative users. These architectures also involve increased complexity as privacy grows and as non-encrypted, encrypted, and homomorphically encrypted solutions are implemented. The latter scheme is the most computationally demanding; however, it provides the highest level of privacy by employing homomorphic encryption, whereby addition and multiplication operations may be performed on encrypted data to produce encrypted outputs, without revealing information about the collaborative agent’s location. The proposed privacy-preserving cooperative position schemes are shown to provide the same results as their non-privacy-preserving counterparts, while providing privacy guarantees. Based on this analysis, some of the proposed solutions can be considered for real-time applications, while homomorphic encryption is a valid solution for latency-tolerant applications. Advances in computing power will increase their overall usability in the near future.
The issue of user privacy in the context of collaborative positioning is addressed in this work, wherein information is passed between and processed by multiple cooperative agents, with the goal of achieving high levels of positioning accuracy. In particular, this paper discusses three privacy-preserving schemes in the context of differential global navigation satellite system (GNSS)-based and GNSS-based cooperative positioning methods. The discussed architectures provide the same positioning results, while yielding different levels of privacy to the cooperative users. These architectures also involve increased complexity as privacy grows and as non-encrypted, encrypted, and homomorphically encrypted solutions are implemented. The latter scheme is the most computationally demanding; however, it provides the highest level of privacy by employing homomorphic encryption, whereby addition and multiplication operations may be performed on encrypted data to produce encrypted outputs, without revealing information about the collaborative agent’s location. The proposed privacy-preserving cooperative position schemes are shown to provide the same results as their non-privacy-preserving counterparts, while providing privacy guarantees. Based on this analysis, some of the proposed solutions can be considered for real-time applications, while homomorphic encryption is a valid solution for latency-tolerant applications. Advances in computing power will increase their overall usability in the near future.