White-Box Cryptography Watermarking for Content Traceability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In the context of white-box cryptography, existing solutions fail to effectively secure digital content against attackers who have complete control over the hardware and software, as they can modify and reverse-engineer cryptographic functions, leading to unauthorized access and distribution of protected content.

Innovation Solution

The implementation of a cryptographic encryption operation that includes a substitution function and a watermarking function, where the watermarking function produces a unique output when a specific input value is detected, allowing content providers to trace illegitimate copies by integrating the watermark into the decryption algorithm or encryption process, making it stealthy and difficult to detect.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If white-box cryptography is used to provide complete control over hardware and software, then security against reverse engineering is improved, but the ability to trace illegitimate copies and prevent unauthorized distribution deteriorates

Engineering Contradiction:
Improvesecurity against reverse engineeringVSAvoidability to trace illegitimate copies
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

A watermarking function is introduced as an intermediary component between the cryptographic function and the output. This watermarking function embeds unique identifiers or traces into the output data without affecting the cryptographic security. The watermark acts as a mediator that carries identification information while the cryptographic function maintains its security properties against reverse engineering.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The cryptographic system is segmented into distinct functional components: the cryptographic function for security and the watermarking function for identification. By separating these functions, the patent achieves both security against reverse engineering and the ability to trace illegitimate copies through the watermark component.

Inventive Principle:
Principle #1Segmentation

2Reliability

If cryptographic functions are obfuscated to prevent reverse engineering, then protection of secure information is improved, but detection and measurement of the watermark becomes more difficult

Engineering Contradiction:
Improveprotection of secure informationVSAvoiddetection of watermark
Core Design Contradiction:
ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The watermarking function applies local modifications to specific portions of the output data rather than altering the entire output. This allows the watermark to be embedded in a localized manner that does not affect the overall cryptographic obfuscation, maintaining security while enabling detection through targeted analysis of the watermarked portions.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3226164B1Watermarking input and output of a white-box implementation
Publication Date: 2020.11.25 NXP BV
  • EP3226164B1 patent drawingFigure 1~2
  • EP3226164B1 patent drawingFigure 3~4
  • EP3226164B1 patent drawingFigure 5~6

AI summary

A method of mapping an input message to an output message by a keyed cryptographic encryption operation, wherein the keyed cryptographic encryption operation includes a first round, including: performing a substitution function on a first portion of the input message to produce an output, wherein the substitution function incorporates a portion of a cryptographic key; and performing a watermarking function on the output, wherein the watermarking function produces a watermark output when the first input portion has a specific predetermined value, wherein the watermark output uniquely identifies the keyed cryptographic encryption operation.