Voice Based User-Device Physical Unclonable Functions for Mobile Device Authentication

Abstract

In this paper, a novel voice based User-Device (UD-) physical unclonable function (PUF) is demonstrated. In traditional PUFs, variability of challenge-response pairs (CRPs) only comes from physical randomness of silicon. Recently, a new type of PUF, touch screen based UD-PUF is proposed, which entangles human user biometric variability with the silicon biometric. Any silicon based mobile device sensor which is a UI element can potentially seed such a UD-PUF. Multiple UD-PUFs based on different, orthogonal sensors enhance robustness. If one UD-PUF behaves poorly in certain environmental conditions, another one might behave well. In the voice UD-PUF, the challenge is a single word chosen by the user. The user speaks the challenge word into the microphone of the mobile device. The speech has natural human biometric variability. The raw microphone output data of the analog to digital converter (ADC) also reflects the silicon variability. This voice microphone data sequence can be quantized into a binary sequence leading to a PUF - a physical randomness derived, unclonable function. To ensure reproducibility, a background noise reduction algorithm is applied on raw voice data sequence. Both variability and reproducibility of this voice UD-PUF are evaluated. We achieve 4.74% false positive rate and 15.56% false negative rate in reproducibility by using Pixel-matching recognition algorithm. For variability, we show 60+ bits Hamming distance, on average, between 128 bits binary responses of different (user, device, challenge) combinations. We also assess the pseudorandom number generation properties of the voice UD-PUF by putting its binary responses through UMontreal TESTU01 suite of tests. The best voice UD-PUF algorithms passed all but 3-6 of the 26 randomness tests.