Hybrid Boolean Networks as Physically Unclonable Functions

We introduce a Physically Unclonable Function (PUF) based on an ultra-fast chaotic network known as Hybrid Boolean Network (HBN) implemented field programmable gate array. The network, consisting of N coupled asynchronous logic gates displaying dynamics the sub-nanosecond time scale, acts `digital fingerprint' by amplifying small manufacturing variations during period transient chaos. In contrast to other PUF designs, we use both N-bits per challenge and obtain response considering challenges be initial states N-node responses captured subsequent transient. find that presence chaos amplifies frozen-in randomness due differences extractable entropy is approximately 50% maximum N2 N bits. uniqueness reliability metrics μ inter = 0.40±0.01 xmlns:xlink="http://www.w3.org/1999/xlink">intra 0.05±0.00, respectively, for N=256 network. These correspond expected Hamming distance 102.4 bits response. Moreover, simple cherry-picking scheme discards noisy yields <; 0.01 while still retaining ~ 200 bits/response (corresponding 80 bits/response). addition characterizing reliability, demonstrate super-exponential scaling in up N=512 PUFmeter, recent analysis tool, unable model our PUF. Finally, characterize temperature variation HBN-PUF propose future improvements.