A True Random Number Generator Using Threshold‐Switching‐Based Memristors in an Efficient Circuit Design

Abstract

A true random number generator (TRNG) based on the stochastic delay and relaxation times of the threshold switching (TS) behavior in a Pt/HfO2/TiN memristor is proposed. The stochasticities of this device are attributed to its electron trapping and detrapping processes. This electronic‐switching‐based memristor exhibits several advantages, such as low power consumption and high reliability. A new circuit is designed to improve the simplicity, miniaturization, and lifetime of TRNG. The bitstreams collected from this TRNG pass the National Institute of Standards and Technology randomness tests without postprocessing, verifying the feasibility of adopting the memristor in hardware security applications. The bit generation rate in this work is sufficient for encryption applications requiring low power and low speed.

Stochastic Spiking Behavior in Neuromorphic Networks Enables True Random Number Generation

“…A range of physical variables and nanoscale devices have previously been employed in TRNGs, as summarized in Section III of the Supporting Information. While impressive performance has been achieved, existing approaches suffer from various drawbacks (such as lack of compatibility with CMOS technologies, scalability, reliability, low bit rates, and low technology readiness levels) , and alternative technologies are still required. In many cases, additional post-processing steps are required to remove intrinsic bias in the underlying physical phenomena, that is, in order to generate random numbers of sufficient quality for applications.…”

A High‐Speed True Random Number Generator Based on a CuxTe1−x Diffusive Memristor

“…The bit generation rate was 32 kb s −1 , which is the fastest speed among the reported volatile‐memristor‐based TRNGs, as shown in Table 1 . [ 24,25,27 ] In this work, the possibility of speed enhancement was experimentally proved through two ways: circuit modification and device engineering. The speed was improved with the combination of NFSR.…”

Low-Power True Random Number Generator Based on Randomly Distributed Carbon Nanotube Networks

“…Subsequently, more advanced TRNGs have been proposed based on the stochastic delay or relaxation time of volatile memristors which have passed all the NIST tests without any postprocessing steps [7]- [9]. Nevertheless, the driving current of prior volatile memristors exceeded 1 μA [7], which lead to inevitably high TRNG power consumption, while an additional peripheral circuit (e.g., a nonlinear feedback shift register circuit [8], [9]) is required to achieve a high bit generation rate. Furthermore, the high endurance performance (>10 10 ) required from the TRNG cannot yet be fully guaranteed using volatile memristors [10].…”