Periodic Reporting for period 1 - entropIC (Anticounterfeiting solution based on integrated circuits implementing high-entropy physical unclonable functions)
Okres sprawozdawczy: 2024-07-01 do 2025-06-30
The EntropIC project aims to advance the technology and develop the business of a breakthrough comprehensive anti-counterfeiting and identity management solution based on proprietary silicon integrated circuits implementing high-entropy physical unclonable functions (HE-PUFs).
Physical Unclonable Functions (PUFs) are silicon integrated circuits that leverage random variations in the fabrication process to implement a unique function that is unknown to the observer and the manufacturer, and extremely difficult to determine experimentally. From a behavioral viewpoint, PUFs are circuits that respond to digital input strings (challenges) with perfectly repeatable digital output strings (responses), where the challenge-response mapping (the unclonable function) is fully repeatable with aging and with varying environmental conditions. As of now, the practical adoption of PUFs has been restricted by their limited entropy, which makes them vulnerable to cyber attacks (e.g. brute-force and AI-based attacks).
In addition to brute-force attacks, weak PUFs are vulnerable to side-channel attacks that can obtain the CRPs, e.g. through power side-channel modeling because the CRP state space is small. Machine learning attacks have been successfully demonstrated even on “strong” PUFs if the attacker gets enough valid CRPs using advanced methods such as approximate attacks.
Physical Unclonable Functions (PUFs) are silicon integrated circuits that leverage random variations in the fabrication process to implement a unique function that is unknown to the observer and the manufacturer, and extremely difficult to determine experimentally. From a behavioral viewpoint, PUFs are circuits that respond to digital input strings (challenges) with perfectly repeatable digital output strings (responses), where the challenge-response mapping (the unclonable function) is fully repeatable with aging and with varying environmental conditions. As of now, the practical adoption of PUFs has been restricted by their limited entropy, which makes them vulnerable to cyber attacks (e.g. brute-force and AI-based attacks).
In addition to brute-force attacks, weak PUFs are vulnerable to side-channel attacks that can obtain the CRPs, e.g. through power side-channel modeling because the CRP state space is small. Machine learning attacks have been successfully demonstrated even on “strong” PUFs if the attacker gets enough valid CRPs using advanced methods such as approximate attacks.
The first year of the EntropIC project has been marked by remarkable progress and groundbreaking achievements towards the overall objective of demonstrating the technological and business viability of comprehensive anti-counterfeiting and identity management solution based on proprietary silicon integrated circuits implementing high-entropy physical unclonable functions (HE-PUFs).
We are confident that our team has worked tirelessly to advance the field of anti-counterfeiting and identity management solutions, and the results represent a significant advancement with respect to the state of the art.
One of the most significant milestones reached this year is the successful definition, design, and fabrication of a novel high-entropy silicon-based strong Physical Unclonable Function (PUF). This innovative PUF, which boasts an impressive entropy of over 8 Mb (we promised 1 Mb for the first generation!), has been optimized for testing and debugging, and we are waiting for the chips from foundry to start experimental validation. The resilience of this two-stage PUF architecture to machine learning attacks has been thoroughly tested, and the results have been outstanding. This achievement not only demonstrates the technological viability of our solution but also sets a new standard in the field (we submitted a dedicated patent application).
We are confident that our team has worked tirelessly to advance the field of anti-counterfeiting and identity management solutions, and the results represent a significant advancement with respect to the state of the art.
One of the most significant milestones reached this year is the successful definition, design, and fabrication of a novel high-entropy silicon-based strong Physical Unclonable Function (PUF). This innovative PUF, which boasts an impressive entropy of over 8 Mb (we promised 1 Mb for the first generation!), has been optimized for testing and debugging, and we are waiting for the chips from foundry to start experimental validation. The resilience of this two-stage PUF architecture to machine learning attacks has been thoroughly tested, and the results have been outstanding. This achievement not only demonstrates the technological viability of our solution but also sets a new standard in the field (we submitted a dedicated patent application).
The design and the development of the two-stage high-entropy PUF is a technical breakthrough and a major achievement of the project. Based on the new design, we are developing the complete authentication infrastructure