Secure Two-Party Computation for Fingerprinting-based Indoor Localization

With the increasing demand for privacy-preserving localization, how to protect the private information of customers (e.g., location) and service providers (e.g., database) is worthy of great attention in the process of realizing indoor positioning. Based on the two-party computation (TPC) protocol, this paper studies and optimizes the indoor localization technology which uses the received signal strength (RSS) of Wi-Fi access point, for the sake of privacy protection. The Paillier encryption algorithm is first used to encrypt the RSS values of the client, in which we propose to optimize the algorithm by predictive calculation and logic optimization, to reduce the communication overhead. Then, the DGK protocol is proposed to calculate the minimum distance in the ciphertext field on the server side. Simulation results show that the proposed algorithm can protect not only the location privacy of the client, but also the database security of the service provider, with a reduced computation cost.