Real-Time Movement Trajectory Anonymization via Dynamic Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anonymization techniques for movement trajectories are ineffective in real-time environments, leading to excessive abstraction of positioning data and loss of information utility, particularly when k-anonymity is not maintained across varying time intervals and geographical changes.
Innovation Solution
An information protection device that includes a movement trajectory storage unit, an anonymous information storage unit, an increment abstraction unit, and a partitioning unit, which generates and partitions anonymous information in real-time to ensure k-anonymity and prevent excessive abstraction, using dynamic reconstruction and merging to maintain data utility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing anonymization techniques are applied to movement trajectories in real-time environments, then anonymity is achieved, but excessive abstraction of positioning data occurs leading to loss of information utility
Solution Approach 1:
The patent applies dynamics by making the anonymization process adaptive to real-time conditions. The system dynamically adjusts the anonymization level based on the current number of users, their distribution patterns, and movement characteristics. This allows the anonymization to maintain k-anonymity while minimizing unnecessary abstraction, thereby preserving information utility in real-time environments.
Solution Approach 2:
The system changes parameters such as the anonymization granularity and grouping criteria based on real-time data characteristics. By adjusting these parameters dynamically, the system achieves effective anonymity without excessive abstraction, resolving the contradiction between anonymity reliability and information utility.
2Reliability
If k-anonymity is enforced across varying time intervals and geographical changes, then privacy protection is improved, but data utility and accuracy deteriorate
Solution Approach 1:
The patent applies local quality by implementing different anonymization strategies for different geographical locations and time intervals. Instead of applying a uniform anonymization level across all data, the system adjusts the anonymization granularity based on local characteristics such as user density, movement patterns, and sensitivity of locations. This ensures adequate privacy protection while preserving data utility in each local context.
Solution Approach 2:
The system dynamically adapts the anonymization level to varying time intervals and geographical changes. By making the anonymization process responsive to real-time conditions, the system maintains appropriate privacy protection without unnecessarily reducing data utility, resolving the contradiction between privacy protection and data utility.
3Reliability
If positioning data is heavily abstracted to ensure anonymity, then anonymity metric is improved, but the accuracy and usefulness of the data for analysis decreases
Solution Approach 1:
The system applies partial anonymization by abstracting only the necessary portions of positioning data to achieve k-anonymity, rather than applying excessive abstraction to all data. This selective approach maintains the anonymity metric while preserving the accuracy and usefulness of the data for analysis, resolving the contradiction between anonymity and data accuracy.
Data Source
AI summary
When position information is added in a real-time fashion, anonymity of movement trajectories is ensured and the degree of abstraction of positioning data included in the position information is prevented from becoming too high. A group of anonymous information with a second positioning time preceding a first positioning time is partitioned into two or more groups so that the anonymous information with the first positioning time satisfies a predetermined anonymity metric and so that a degree of abstraction of the abstracted positioning data in the anonymous information becomes lower than a predetermined standard value, anonymous information including positioning data with the first positioning time abstracted by the partitioned groups is generated, and the generated anonymous information is stored in an anonymous information storage unit in association with the anonymous information with the second positioning time.