PUF Key Entropy Enhancement via Random Number Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-volatile memory-based physical unclonable functions (PUFs) face challenges with high bit error rates due to resistance drift in memory cells, particularly at elevated temperatures, which affects the reliability of generated keys used in security protocols for IoT devices and other applications.
Innovation Solution
A combination of pseudo-random number generators and PUF circuits is used to enhance key entropy by logically combining PUF keys with random numbers, employing hash functions or XOR operations to produce stable, low-error-rate enhanced keys, which are stored in non-volatile memory cells, ensuring secure and reliable key management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If non-volatile memory cells are used for PUF key generation, then key storage capability is improved, but bit error rate increases due to resistance drift
Solution Approach 1:
The patent combines PUF key generation with random number generation in a unified security key generation system. The PUF circuit provides a baseline unique identifier while the RNG adds entropy, and their combination through logical operations produces enhanced security keys that leverage both sources of randomness to overcome the limitations of PUF alone
Solution Approach 2:
The patent applies parameter changes by using hash functions and logical operations (XOR, AND, OR) to transform the PUF output and combine it with random numbers. This transformation process converts the high-error-rate PUF key into a lower-error-rate enhanced key by distributing and diffusing the error characteristics across multiple bits
2Measurement precision
If PUF circuits are used for key generation, then unique identifier capability is improved, but key entropy is insufficient for high security requirements
Solution Approach 1:
The patent merges PUF circuit output with random number generator output to create enhanced security keys. The PUF provides unique device identification based on manufacturing variations, while the RNG contributes additional entropy, and their combination through logical operations produces keys with both uniqueness and high entropy required for strong security
3Reliability
If error correcting codes are used to improve PUF reliability, then bit error rate is reduced, but device complexity increases
Solution Approach 1:
The patent introduces random numbers as an intermediary element that mediates between the PUF output and the final security key. Instead of directly correcting PUF errors through complex coding, the system uses random numbers combined with logical operations to mask and distribute errors, achieving reliability improvement with simpler circuitry
Data Source
Figure 1
Figure 2~3
Figure 4A~4E
AI summary
A device which can be implemented on a single packaged integrated circuit or a multichip includes a plurality of non-volatile memory cells, and logic to use a physical unclonable function to produce an initial key and to store the initial key in a set of non-volatile memory cells in the plurality of non-volatile memory cells. The device can include logic to use a random number generator to generate a random number, and logic to combine the initial key and the random number to produce an enhanced key. The physical unclonable function can use entropy derived from non-volatile memory cells in the plurality of non-volatile memory cells to produce the initial key. Logic is described to disable changes to data in the set of non-volatile memory cells, and thereby freeze the key after it is stored in the set.