Digital Fingerprinting via Principal Component Marking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital fingerprinting systems are vulnerable to collusion attacks, which allow malicious users to create copies devoid of fingerprints or that incriminate innocent users, limiting their effectiveness and scalability, and traditional watermarks based on random patterns are not suitable for evidence due to potential interference and low information capacity.
Innovation Solution
The method involves transforming digital signals into a discrete orthonormal basis, ranking and selecting principal components, altering them with a marking angle, and combining them with unmarked components to create a fingerprinted signal, using pseudo-random noise arrays for synchronization and concealing additional data, which enhances collusion resistance and information capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If traditional random pattern watermarks are used for digital fingerprinting, then the implementation is simple, but the system is vulnerable to collusion attacks and has low information capacity
Solution Approach 1:
The patent transforms the watermarking approach by changing from random patterns to structured sequences with specific mathematical properties (low cross-correlation, good autocorrelation). This parameter change in the sequence structure provides collusion resistance while maintaining implementation feasibility through systematic generation methods.
Solution Approach 2:
The patent combines multiple sequences with complementary properties into a composite watermarking system. By using sequences with low cross-correlation and good autocorrelation properties together, the system achieves both collusion resistance and high information capacity, overcoming the limitations of single random patterns.
2Loss of information
If multiple patterns are embedded in one file to increase information capacity, then more recipients can be identified, but potential mutual interference occurs reducing reliability
Solution Approach 1:
The patent extracts and utilizes specific mathematical properties (low cross-correlation, good autocorrelation) from structured sequences. By selecting sequences with these extracted properties, multiple patterns can be embedded without mutual interference, as the low cross-correlation ensures patterns remain distinguishable even when combined.
Solution Approach 2:
The patent changes the fundamental parameters of the watermark sequences from random to structured with specific correlation properties. This parameter change enables multiple patterns to coexist in one file without interference, as the structured sequences maintain their individual characteristics even when embedded together.
3Reliability
If digital fingerprints are made robust against processing distortions, then reliability improves, but the fingerprints become more detectable by potential attackers
Solution Approach 1:
The patent applies different properties to different aspects of the watermark system: structured sequences with specific correlation properties provide robustness to distortions, while the imperceptible embedding method ensures attackers cannot easily detect the fingerprints. Each component has optimized local quality for its specific function.
Solution Approach 2:
The patent uses imperceptible watermarking techniques that embed fingerprints in a way that is undetectable to attackers, effectively making the fingerprint system disposable or transient in terms of detectability, while maintaining robustness through the mathematical properties of the sequences.
4Reliability
If the fingerprint detection is done in the presence of the original clip, then robustness requirement is easier to satisfy, but the system becomes more complex requiring original signal storage
Solution Approach 1:
The patent enables the watermarked signal to serve itself by embedding synchronization information and structural markers within the watermark sequences. This self-service capability allows the system to automatically locate and identify fingerprints without requiring external reference to the original signal, reducing system complexity while maintaining robustness.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A device to apply a digital fingerprint to a digital signal comprises a means of intercepting or acquiring a signal, a storage element and a processor for executing computer implemented programme code components in the storage element to effect the methods. The methods include transforming a plurality of signal samples onto a discrete orthonormal basis and ranking the transformed samples according to their magnitude. The n largest principal components of the ranked transformed samples are optionally permuted to generate a re-ordered set of principal components, which are then altered by a marking angle. The marked principal components and unmarked non-principal components are converted and combined and applying an inverse of the transform function to the combined principal and non-principal components to generate a fingerprinted digital signal. Methods to prepare the signal for marking, recover the digital fingerprint and verify the distributor and/or recipients of the signal are also disclosed.