Quotient Ring Transform for Secure Random Data Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deterministic random number generators (DRNGs) face issues such as correlation of successive cryptographic random data, loss of entropy, and vulnerability to attacks due to repetitive values and entropy loss in block ciphers and hashing functions, compromising computer and network security.
Innovation Solution
A device uses a quotient ring transform (QRT) and Galois stepping to generate cryptographic random data from raw random data, maintaining entropy flow by storing raw data in a Galois register and transforming it based on an extraction state, ensuring secure generation of random data for cryptographic and network security processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If deterministic algorithms (block ciphers, hashing functions) are used to generate random data, then the generation process is efficient and deterministic, but the output exhibits correlation between successive values and loses entropy, compromising security
Solution Approach 1:
The patent introduces a Galois register as an intermediary component between the deterministic algorithm output and the final cryptographic random data. The Galois register receives deterministic random data and applies Galois stepping transformations to produce cryptographic random data with improved security properties. This intermediary structure maintains the efficiency of deterministic generation while eliminating correlation and entropy loss through the Galois field operations.
Solution Approach 2:
The patent transforms the output parameters of deterministic algorithms by applying Galois stepping operations in a Galois register. Instead of directly using the deterministic algorithm output, the system changes the parameter representation through finite field arithmetic, specifically using the transformation x → x^2 + x + 1 in GF(2^m). This parameter transformation eliminates correlations and preserves entropy, resolving the security issue while maintaining generation efficiency.
2Ease of manufacture
If block ciphers and hashing functions are used for random data generation, then the process is deterministic and efficient, but the data becomes vulnerable to attacks due to repetitive values and entropy loss
Solution Approach 1:
The Galois register serves as a protective intermediary layer between the simple deterministic algorithms and the cryptographic application. This intermediary applies Galois field transformations that are computationally simple yet effectively prevent cryptographic attacks by eliminating repetitive patterns and preserving entropy. The solution maintains implementation simplicity while adding a security layer that counters harmful factors.
Solution Approach 2:
The patent converts the deterministic nature of block ciphers and hashing functions, which normally causes correlation and security vulnerabilities, into a benefit by using it to efficiently fill the Galois register. The Galois stepping then transforms this deterministic input into unpredictable cryptographic random data. The potential harm of determinism is converted into the benefit of efficient entropy collection, which is then transformed into secure random data.
Data Source
AI summary
A device may store raw random data in a raw random data store. The raw random data may include a first plurality of data strings. The device may generate, using a quotient ring transform (QRT), cryptographic random data based on the raw random data. The cryptographic random data includes a second plurality of data strings that is transformed from the first plurality of data strings based on an extraction state stored in an extraction state store. The device may store the cryptographic random data in a cryptographic random data store and may use the cryptographic random data for various purposes.