What is the photonic Physical Unclonable Function device you are developing and what is its role in the SMILE Action?
A Physical Unclonable Function (PUF) is a physical implementation of a one-way mathematical function. Such a function has the property to map each one of its inputs to a unique output, and by knowing the output, no one can reproduce the input. This feature is very convenient for cryptographic applications. There are a lot of physical media that have such a property, and by stimulating them with the appropriate set of inputs (challenges), some unique outputs (responses) can be generated. PUF shifts the security from the software to the hardware level, and these challenge-response pairs can be used for authentication and data encryption purposes and shift the security software level to the hardware level.
In the context of SMILE action, the PUF device will be installed at the BCPs and will be integrated inside the Smart Gateway and will provide the system with true random numbers. These will be used as seeds for the homomorphic algorithms for data encryption purposes and will also generate memoryless and scalable authentication keys for authenticating the BCP user/tablet. The most significant advantage of this technology is the real-time reproduction of the authentication keys, which means that the authentication keys are reproduced upon request and therefore there is no need to store the user-key in a centralised database permanently. As a result, massive key leakage is prevented. To summarise, SMILE system will offer a unique and enhanced data security integrated into the hardware level by exploiting the innovative technology of p-PUF.
What is the competitive advantage of your device in terms of security compared to already in market solutions?
PUF devices are characterised by four main features: their entropy, anti-tampering capabilities, resilience to adversary attacks and footprint. Currently, the most significant competitors to the photonic PUFs are the electronic PUFs that are already in the market. Their main advantages over p-PUF are that they have a small footprint, lower energy consumption, and cost-effectiveness. These features are significant for applications that require the authentication and data security of small and mobile devices.
On the other side, p-PUFs offer a higher entropy, against their electronic counterparts, due to their increased structural complexity, which provides an exponentially large pool of random numbers/authentication keys, appropriate for application with a large number of users/devices and increased data rates. Also, due to intrinsic sensitivity to minor variations of different mechanical parameters such as tension, stress, temperature, and other, they offer inherent anti-tampering capabilities. Finally, due to their random and their highly complex physical structure they are resilient to machine learning and modelling attacks.
At what stage of development of the device are you currently and what is your goal for the end of the Action?
At the beginning of the Action, PUF device was in an experimental proof of concept level. We developed a working prototype in the laboratory which allowed us to verify its capability to generate random numbers and authentication keys for a wide range of applications. By defining the specifications of the system, based on the SMILE system architecture, we designed the first small footprint prototype of the PUF device which in the next three months will be ready to be installed in a rack for a simulated environment operation.
Our goal, by the end of the SMILE Action, is to develop a standalone miniaturised version of the PUF module that will be integrated inside the Smart Gateway, which in turn will be installed at the Border Crossing Points premises.
Dr Alexandros Markos Fragkos is an RnD Manager in Optoelectronics at