Physical Unclonable Function Seed Stabilization for Secure PRNG
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Pseudo-random number generators (PRNGs) used for cryptographic key generation require secrecy of the seed, as using the same seed results in the same sequence, making it insecure if exposed, and traditional hardware-based generators lack reproducibility and are costly.
Innovation Solution
A method utilizing a Physical Unclonable Function (PUF) to generate a stable response from a seed, which is then used to initialize a PRNG, ensuring the seed does not need to be kept secret and the output cannot be reproduced without the PUF, combining the response with helper data to compensate for noise and ensure reproducibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a pseudo-random number generator is used with a seed to generate arbitrary numbers, then the numbers appear random and are easy to generate, but the seed must be kept secret to prevent reproduction of the sequence
Solution Approach 1:
A physical unclonable function (PUF) is introduced as an intermediary between the seed and the pseudo-random number generator. The PUF takes the seed as input and produces a challenge-response pair where the response is used to initialize the PRNG. This intermediary protects the seed from direct exposure while enabling reproducible number generation.
Solution Approach 2:
The system is segmented into distinct functional components: the PUF module that processes the seed, the challenge-response generation mechanism, and the PRNG module. This segmentation allows the seed to be handled separately from the number generation process, improving security while maintaining ease of use.
2Reliability
If hardware-based random number generators are used to provide truly random numbers, then security is improved, but the system becomes expensive and cumbersome with limited reproducibility
Solution Approach 1:
Instead of using expensive hardware random number generators, the system creates a virtual copy that uses a PUF-based approach. The PUF generates challenges that can be reproduced software-based, eliminating the need for physical hardware while maintaining security properties and enabling full reproducibility through the challenge-response mechanism.
Solution Approach 2:
The patent replaces mechanical/hardware-based random number generation with a software-based system using physical unclonable functions. The PUF leverages physical properties at a fundamental level but implements them through computational challenges and responses, substituting complex hardware with a more manageable software implementation.
3Stability of the object's composition
If the same seed is used to initialize a pseudo-random number generator, then the same sequence of numbers is produced, but the sequence becomes predictable and insecure
Solution Approach 1:
The system performs preliminary action by having the PUF generate a challenge-response pair before the PRNG is initialized. The challenge is derived from the seed through the PUF, and the response is used to seed the PRNG. This preliminary processing ensures that even if the same seed is used, the intermediate challenge-response transformation prevents direct prediction of the number sequence.
Data Source
AI summary
The invention provides a method of generating arbitrary numbers given a seed, characterized by providing a challenge derived from the seed to a physical token, receiving an initial response from the physical token, combining the initial response with helper data associated with the challenge to produce a stable response, and generating the arbitrary numbers using a pseudo-random number generator using the stable response as a seed for the generator. Preferably one or more of these pseudo-random permutations are used as one or more round function(s) in a Feistel block cipher. The generated arbitrary numbers may also be used to create a cryptographic key.