Time sequence position publishing method and system

Abstract

The invention relates to a differential privacy time sequence position publishing method and system based on a personalized privacy policy, and the method comprises the steps: determining an LBS service position domain, carrying out the grid division, dividing the LBS service position domain into a plurality of position points, and obtaining the position points under a grid coordinate system, wherein each grid comprises a position point; constructing an initial undirected graph according to all position points under the grid coordinate system; obtaining a probability state transition matrix of the hidden Markov model according to the historical position data of the user; and disturbing the real position of the user based on the probability state transition matrix and the initial undirected graph in combination with a Laplacian mechanism, and obtaining and publishing a disturbed position. The user is allowed to customize the connection relation between the nodes in the undirected graph in a personalized manner, so that the user can adjust the privacy protection degree according to the privacy requirement of the user. Meanwhile, Laplace noise is introduced, and the time sequence position is published by using the hidden Markov model, so that the security of user position publishing is improved, and the privacy leakage of the real position of a user is avoided.

Description

technical field [0001] The present invention relates to the technical field of time-series position release, in particular to a time-series position release method and system. Background technique [0002] When the attacker has a large amount of background knowledge, the K-anonymous location privacy scheme is usually used to anonymize the user's location. The K-anonymous location privacy scheme samples the trajectory Ti at any time, so that at the sampling time, there are at least k- One trajectory can achieve the effect of privacy protection in the same area as Ti generalization at the corresponding position. The generalization and suppression technology represented by the K-anonymity scheme is difficult to resist the attacker's background knowledge attack. For example, the attacker can infer the anonymized position based on historical information, making the generalization invalid. In addition, the K-anonymity scheme is not flexible enough for specific LBS services (locat...

AI Technical Summary

Problems solved by technology

The generalization and suppression technology represented by the K-anonymity scheme is difficult to resist the background knowledge attack of the attacker. For example, the attacker can infer the anonymized position based on the historical information, making the generalization invalid

In addition, the K-anonymity scheme is not flexible enough for specific LBS services (location-based information services, Location Based Services), and users cannot adjust the specific degree of privacy protection according to their needs.

Images