An Arbiter PUF employing eye-opening oscillation for improved noise suppression

Abstract

Like every integrated circuit, Arbiter-PUFs suffer from any ambience variations such as electrical noise, supply voltage variations and temperature fluctuations. In this work, we show that most bit-errors are related to small phase differences, for which noise dominates the readout bit value. We present an approach eliminating this influence by modifying the arbitration circuit part into an eye-opening oscillator. By utilizing the deadzone of two D-Flip-Flops, the decision about the response is delayed until the phase difference becomes significant enough. With this modification, we could increase the PUFs initial biterror rate of 3.31% to almost zero. We also highlight important design choices for this solution, like the setup-time of the D-Flip-Flops and the minimum number of enforced oscillations.

Reliability of Ring Oscillator PUFs with Reduced Helper Data

“…We do not observe significant changes in performances between the square-root boundary and the constant boundary. Hence we advocate to use the simpler constant boundary as introduced by Herkle et al [21]. It is interesting to observe that despite higher temperature should increase the KER because of larger noise variance the performances are barely affected by the temperature.…”

“…A feedback-based mode of operation monitors each bits independently and decides when it is reliable enough in order to minimize the global Key Error Rate (KER). This formalizes the eye-opening concept used in Herkle et al [21]. 3.…”

“…Their construction differs from classical constructions in that it is not parametric but relies on random but permanent connection in the hardware layout. -Herkle et al [21] present an eye-opening oscillator for arbiter PUFs, which exploit the dead-zone of the arbiter PUF to decide whether a bit is reliable or not, and automatically request new oscillations accordingly. Monte-Carlo simulation with transient noise and 50 repetitions yields an expected Bit Error Rate (BER) of 9.2 • 10 −5 ± 7.7 • 10 −4 .…”

“…For that we proceed in two steps. First we improve the reliability of the LPUF [6] by adaptively controlling the number of required oscillations (as hinted in [21]). Second, we investigate the performances of lightweight helper data (less than one byte).…”