Cpir‑v Nearest Neighbor Privacy Preserving Query Method Based on Local Superset

Abstract

The invention discloses a CPIR‑V nearest neighbor privacy protection query method based on local supersets. The invention first searches for the relationship between potential nearest neighbor point sets in the grid, then establishes the mapping relationship between grids, and then finds The grid relationship compresses the potential nearest neighbor storage matrix, thereby achieving the purpose of reducing the amount of calculation and communication cost. The amount of data required to be calculated, the communication cost, and the amount of client calculation are all reduced.

Description

technical field [0001] The invention relates to a query method, in particular to a local superset-based CPIR-V (SCPIR-V) nearest neighbor privacy protection query method. Background technique [0002] The existing CPIR-V algorithm realizes the privacy-preserving query of spatial nearest neighbors, but the algorithm has the following two shortcomings. [0003] (1) The computational complexity is O(m·s·r) (where m=sizeof(p i ).P max , s and r are the number of columns and rows respectively, in figure 2 where s=r=5), the CPU cost is relatively large. Because there are many repeated multiplication operations in the matrix, an optimization strategy based on data mining technology is proposed in the literature. This method uses the Apriori algorithm to extract repeated calculations, and saves the calculation results to reduce repeated calculations. Although the amount of calculations after optimization It has been optimized, but the calculation cost of CPU is still high. [0...

AI Technical Summary

Problems solved by technology

[0003] (1) The computational complexity is O(m·s·r) (where m=sizeof(p i ).P max , s and r are the number of columns and rows respectively, in figure 2 In s=r=5), the CPU cost is relatively large

Because there are many repeated multiplication operations in the matrix, an optimization strategy based on data mining technology is proposed in the literature. This method uses the Apriori algorithm to extract repeated calculations, and saves the calculation results to reduce repeated calculations. Although the amount of calculations after optimization It has been optimized, but the calculation cost of the CPU is still high

[0004] (2) The communication complexity is O(k.s+k.m.r) (where s and r are the same as above, k is the binary digit of POI data), and the paper proposes to calculate the optimal value of s and r to reduce the communication complexity , but it cannot be applied to the spatial nearest neighbor query. The query results are compressed by using standard compression techniques, but the transmission of the query results still consumes a lot of bandwidth

This is a huge drain on system resources

Images