Achieving Differential Privacy Publishing of Location-Based Statistical Data Using Grid Clustering

Statistical partitioning and publishing is commonly used in location-based big data services to address queries such as the number of points interest, available vehicles, traffic flows, infected patients, etc., within a certain range. Adding noise perturbation statistical according differential privacy model can reduce various risks caused by location leakage while keeping characteristics published data. The traditional methods realize decomposition indexing 2D space from top bottom. However, they easily cause over-partitioning or under-partitioning phenomenon, therefore need multiple times scan. This paper proposes grid clustering protection method for scenarios. We implement statistics units equal-sized grids perform density classification on uniformly distributed discrete wavelet transform. A bottom-up algorithm designed blank uniform same level based neighborhood similarity. Laplacian incorporated into results form statistics. Experimental comparison real-world datasets manifests that proposed this superior other existing partition terms range querying accuracy operating efficiency.