Cross-border navigation trajectory computation based on CKKS fully homomorphic encryption algorithm

Cross-border navigation trajectory computation based on CKKS fully homomorphic encryption algorithm

연구 분야: Cryptography

학회: DEAI '25: Proceedings of the 2nd Guangdong-Hong Kong-Macao Greater Bay Area International Conference on Digital Economy and Artificial Intelligence

At present, due to the concerns of overseas users about the risk of data storage and privacy disclosure, the promotion of BeiDou positioning and navigation services in overseas faces obstacles. One potential way to achieve data privacy by minimizing the users’ risk of concern is to implement privacy computing technologies such as cross-border data access and cross-border computation. Besides policy impact and practical application needs, it is found that now existing Zones A and B will implement different data privacy protection regulations, which can be used as related research and experiment platforms. The CKKS fully homomorphic encryption algorithm is used to encrypt and calculate cross-border navigation data for the above-mentioned application scenarios, meeting users’ query and calculation requirements and keeping privacy at the same time. In this study, running tests on the real BeiDou navigation data set is encrypted and calculated; it generates a vhhicle's driving trajectory map based on the user's query request. This study then optimizes the computation process with data partitioning and parallel computation techniques and reduces the time spent during homomorphic encryption as well as computation. The time taken for the calculation hereby is reduced to around 27% as compared to the original serial calculation process.