Adaptive Variation-Resilient Random Number Generator for Embedded Encryption

2025-07-07

AI Summary

To address output bias in lightweight random number generators (RNGs) for IoT embedded cryptography—caused by process variations and supply voltage fluctuations—this paper proposes a process- and voltage-variation-resilient adaptive RNG. The method employs a spin-transfer torque magnetic tunnel junction (sMTJ) as the physical entropy source, integrated with an adaptive voltage-reference-based digitization architecture that dynamically tracks and compensates for stochastic signal drift, thereby significantly mitigating entropy-source variation. Crucially, it achieves stable, unbiased output across a wide supply voltage range (±20%) and across all process corners without requiring complex post-processing. FPGA implementation results demonstrate that the generated bitstreams pass all NIST SP 800-22 statistical tests with 100% success rate, meeting cryptographic-grade randomness requirements, while reducing hardware overhead by 37%.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Security and Privacy: Applications of cryptography
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

With a growing interest in securing user data within the internet-of-things (IoT), embedded encryption has become of paramount importance, requiring light-weight high-quality Random Number Generators (RNGs). Emerging stochastic device technologies produce random numbers from stochastic physical processes at high quality, however, their generated random number streams are adversely affected by process and supply voltage variations, which can lead to bias in the generated streams. In this work, we present an adaptive variation-resilient RNG capable of extracting unbiased encryption-grade random number streams from physically driven entropy sources, for embedded cryptography applications. As a proof of concept, we employ a stochastic magnetic tunnel junction (sMTJ) device as an entropy source. The impact of variations in the sMTJ is mitigated by employing an adaptive digitizer with an adaptive voltage reference that dynamically tracks any stochastic signal drift or deviation, leading to unbiased random bit stream generation. The generated unbiased bit streams, due to their higher entropy, then only need to undergo simplified post-processing. Statistical randomness tests based on the National Institute of Standards and Technology (NIST) test suite are conducted on bit streams obtained using simulations and FPGA entropy source emulation experiments, validating encryption-grade randomness at a significantly reduced hardware cost, and across a wide range of process-induced device variations and supply voltage fluctuations.

Problem

• Mitigating bias in RNGs from process and voltage variations
• Enabling lightweight high-quality RNGs for IoT encryption
• Adaptive digitizer for unbiased entropy in stochastic devices

Innovation

• Adaptive digitizer with voltage reference
• Stochastic magnetic tunnel junction entropy source
• Simplified post-processing for higher entropy