MSET-Based Time-Series Data Provenance Certification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to effectively and efficiently certify the provenance of time-series data stored in time-series databases, which is crucial for ensuring data integrity, debugging, compliance with regulations, and data privacy agreements.
Innovation Solution
The system employs the Multivariate State Estimation Technique (MSET) to estimate and reconstitute time-series data, comparing the reconstituted data with original data to certify provenance, using analytical resampling, clustering, and reversible MSET computations, along with sensor operability flags, to validate the integrity of the data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional data storage methods are used in time-series databases, then data storage capacity is maintained, but data provenance certification capability is lost
Solution Approach 1:
The patent introduces MSET estimates as an intermediary mechanism between original sensor data and provenance certification. These estimates act as a mediator that captures the essential relationships and patterns in the data, enabling provenance verification without requiring storage or processing of all original raw data, thus reducing system complexity while maintaining certification capability
Solution Approach 2:
The patent creates a computational copy of the original data relationships through MSET estimates. Instead of storing and processing the entire original dataset for provenance verification, the system generates compressed representational copies (MSET estimates) that preserve the essential data characteristics and enable efficient provenance certification
2Reliability
If complete time-series data is stored for provenance verification, then data integrity is maintained, but storage requirements and processing time increase
Solution Approach 1:
The patent extracts only the essential provenance-critical information from complete time-series data through MSET estimation. By separating and retaining only the necessary data characteristics (MSET estimates) needed for provenance verification, the system achieves efficient verification without processing or storing the entire original dataset, thus reducing processing time while maintaining data integrity
Solution Approach 2:
The patent applies partial action by performing provenance verification on MSET estimates rather than complete original datasets. This partial verification approach is sufficient for proving data provenance without requiring exhaustive processing of all original data points, significantly reducing processing time while maintaining verification effectiveness
3Productivity
If MSET estimates are computed and stored, then provenance certification efficiency is improved, but computational overhead increases
Solution Approach 1:
The patent performs MSET estimate computation as a preliminary action during data ingestion or batch processing, before provenance certification is needed. By pre-computing these estimates and storing them alongside the original data, the system eliminates the need for complex real-time computations during verification, improving certification efficiency while distributing computational overhead across less critical time periods
Data Source
AI summary
The disclosed embodiments relate to a system that certifies provenance of time-series data in a time-series database. During operation, the system retrieves time-series data from the time-series database, wherein the time-series data comprises a sequence of observations comprising sensor readings for each signal in a set of signals. The system also retrieves multivariate state estimation technique (MSET) estimates, which were computed for the time-series data, from the time-series database. Next, the system performs a reverse MSET computation to produce reconstituted time-series data from the MSET estimates. The system then compares the reconstituted time-series data with the time-series data. If the reconstituted time-series data matches the original time-series data, the system certifies provenance for the time-series data.