PUF Signature Regeneration Using Trellis Decoding and Compact Helper Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Physically unclonable function (PUF) circuits face challenges in maintaining consistency due to device aging, leading to instability and potential authentication failures, with existing helper data being large and potentially revealing sensitive information.
Innovation Solution
A circuit arrangement and method that translate normally distributed raw sample values into uniformly distributed binned values, generating helper data with centering values and parity bits to reduce the quantity of helper data and minimize information leakage, using a trellis decoder to regenerate a PUF signature and validate its validity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional error-correcting code helper data is used to correct raw PUF data, then PUF authentication reliability is improved, but helper data size increases creating storage issues and information leakage
Solution Approach 1:
The helper data is segmented into two distinct components: (1) a compact codebook containing pre-computed correction values for common PUF variations, and (2) a small set of parameters describing the specific device's characteristics. This segmentation allows the system to provide comprehensive error correction while keeping the total helper data size minimal, thus resolving the contradiction between reliability and storage requirements
Solution Approach 2:
The invention extracts and separates the essential correction information from the full error-correcting code structure. By identifying and storing only the critical parameters that characterize device variations, rather than storing complete correction tables, the system achieves effective error correction with dramatically reduced helper data size, preventing information leakage while maintaining authentication reliability
2Stability of the object's composition
If helper data is increased to improve PUF stability over device aging, then authentication consistency is improved, but information security deteriorates due to potential leakage
Solution Approach 1:
The helper data structure uses local quality by providing different types of information in different parts of the helper data: the codebook contains general correction patterns applicable to many devices, while the parameter set contains specific local adjustments for individual device characteristics. This localized approach provides precise correction for each device without requiring large amounts of sensitive information to be stored, thus maintaining stability while minimizing information leakage
Data Source
AI summary
A physically unclonable function includes a circuit that translates a normally distributed sequence of raw sample into a sequence of uniformly distributed binned values across sub-bins of bins. Helper circuitry generates centering values and parity bits based on binned values generated during registration. Each centering value is associated with a raw sample value corresponding to a binned value and indicates an offset of a sub-bin in one of the bins. A distance calculator generates a set of distances from each raw sample value based on the centering value associated with the raw sample value. Each distance indicates a difference between the respective raw sample value and a raw sample value equivalent to a midpoint of a sub-bin offset by the associated centering value in a bin. A trellis decoder generates a PUF signature based on the candidate symbols, sets of distances, and parity bits.