Neuromorphic chips generate random numbers using brain-like networks
Neuromorphic Pseudo-Random Number Generators with a Low Power Hardware Implementation
Neural and Evolutionary Computing
Summary
Generating random numbers usually requires a balance between quality, speed, and power use. The authors looked at how the brain creates unpredictable patterns and built computer models that mimic these brain-like networks. They made a small hardware device that uses these models to produce random numbers while using very little power. The random numbers generated passed standard quality tests both in software and hardware versions.
What this means in practice
- •For embedded system designers: Integrate low power neuromorphic random number generators into devices requiring secure and efficient number generation.
- •For hardware security engineers: Build hardware modules with brain-inspired networks to enhance unpredictability in cryptographic key generation.$Commercial implications: This paper enables the creation of power-efficient hardware RNGs for cryptographic products where security and energy use are critical.
Authors
Jafar Shamsi, Navid Akbari, Sonia Sennik, Aaron Gruber, Wilten Nicola
Abstract
Pseudo-random number generation often requires trade-offs among quality, power consumption, and bandwidth to produce unpredictable sequences of numbers. The brain, on the other hand, efficiently generates unpredictable output complex network dynamics occurring in a high-dimensional state. This state, which is hypothesized to be chaotic, relies on the balance between excitation and inhibition. Here, we investigated if computational models of these chaotic balanced states can be harnessed for Neuromorphic Pseudo-Random Number Generators (NPRNGs) in low power hardware. We successfully constructed a balanced spiking neural network model consisting of leaky-integrate-and-fire neurons that could be readily implemented in low power FPGAs and used as a NPRNG. The prototyped NPRNG consumed 3.24 mW during operation and produced pseudo-random numbers at 120kbps. In both hardware and software instantiations, NPRNGs produce high-quality random numbers as validated by standard metrics for testing RNG quality.