A Density-Based Space Filling Curve for Location Privacy-Preserving

Location-aware devices and cloud computing services have triggered the widespread development of location-based applications. However, such applications require users to disclose their actual locations, which raises serious privacy concerns. As a kind of spatial transformation method, the standard Hilbert curve (SHC) is widely used in privacy preserving for spatial data. But there are still many limitations for SHC, such as high CPU cost, using the unified partition granularity and inability to support partial region authorization, etc. In this paper, we present a density-based space filling curve (DSC) for spatial data transformation, which partition the space based on the density of points of interest (POI) and supports partial authorization of the space, and need relatively low CPU cost. Our experiments show that the security risk of DSC is lower than that of SHC, and DSC provides better index building and range query processing performance than SHC.