Title: Location big data differential privacy dynamic partition release method

Authors: Yan Yan; Lian Xiu Zhang; Bing Qian Wang; Xin Gao

Addresses: School of Computer and Communication, Lanzhou University of Technology, Lanzhou, 730050, China ' School of Computer and Communication, Lanzhou University of Technology, Lanzhou, 730050, China ' School of Computer and Communication, Lanzhou University of Technology, Lanzhou, 730050, China ' School of Computer and Communication, Lanzhou University of Technology, Lanzhou, 730050, China

Abstract: Aiming at the privacy protection requirements in real-time statistical publishing process of location big data, a dynamic partition method is proposed based on differential privacy mechanism. The temporal redundancy between adjacent data snapshots has been eliminated by sampling and differential processing of dynamic location big data, and the spatial redundancy of location big data has been reduced by adaptive density meshing and uniformity heuristic quadtree partitioning. Differential privacy protection has been realised by adjusting partition structures of the current dataset on the spatial structure of previous moment and adding Laplace noise. Experiments carried out on the cloud computing platform and real location big datasets show that the proposed algorithm can meet the dynamic partition release requirements of real-time location big data, and the query precision of single-released location big data is better than other similar methods.

Keywords: location big data; LBD; dynamic partition release; DPR; differential privacy; temporal redundancy; spatial redundancy.

DOI: 10.1504/IJSN.2020.106505

International Journal of Security and Networks, 2020 Vol.15 No.1, pp.25 - 35

Received: 28 Mar 2019
Accepted: 18 Apr 2019
Published online: 09 Apr 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article