Distributed Anomaly Detection Using Secret-Shared Data Transforms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly detection systems in distributed networks fail to identify both local and global anomalies while preserving data privacy, as they often require transmitting raw data to a central node, potentially violating privacy principles and missing anomalies due to sophisticated cyberattacks.
Innovation Solution
A method employing time series discord analysis with secret shared transformation matrices is used to detect anomalies at both local and global levels without revealing original data, by generating a shared transformation matrix across client nodes to transform data before transmission, allowing anomaly detection at a central node without compromising privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If raw data is transmitted to a central node for anomaly detection, then anomaly detection capability is improved, but data privacy is compromised
Solution Approach 1:
The patent applies preliminary action by transforming the data at each peer node before transmission. The data is pre-processed through cryptographic transformations and dimensionality reduction techniques, so that when the data reaches the central node, it is already in a form that preserves privacy while maintaining anomaly detection capability. This resolves the contradiction by preparing the data in advance to simultaneously achieve both privacy protection and effective anomaly detection.
2Loss of information
If data is transformed before transmission to preserve privacy, then data privacy is improved, but anomaly detection accuracy may deteriorate
Solution Approach 1:
The patent applies parameter changes by carefully selecting and adjusting the transformation parameters to maintain the essential characteristics of the data needed for anomaly detection. The dimensionality reduction and cryptographic transformations are designed with specific parameters that preserve the statistical properties and patterns necessary for detecting anomalies, while still achieving privacy protection. This resolves the contradiction by changing parameters in a controlled manner that balances privacy and detection accuracy.
3Ease of operation
If conventional anomaly detection is used without transformation, then detection simplicity is improved, but privacy protection deteriorates
Solution Approach 1:
The patent introduces an intermediary mechanism in the form of transformation matrices and dimensionality reduction techniques that act as mediators between the raw data and the anomaly detection process. These intermediaries transform the data into a privacy-preserving representation while maintaining the essential patterns needed for detection. The central node works with transformed data rather than raw data, simplifying the privacy protection aspect while maintaining detection capability through the intermediary transformation layer.
Data Source
AI summary
One example method may be performed in a distributed environment that includes a group of nodes configured to communicate with each other, and to communicate with a central node of the distributed environment. The example method includes performing, by two or more of the nodes, the operations of generating a public/private key pair, generating a secret shared key using the respective public keys of the other nodes, using the secret shared key to generate isometric transformations of data of the node, and transmitting the transformed data to the central node.