HYPERLOCK: In-Memory Hyperdimensional Encryption in Memristor Crossbar Array

TL;DR

HYPERLOCK introduces a memristor-based cryptography system leveraging stochastic properties and neural networks, enabling secure, high-accuracy image and text encryption despite circuit non-idealities.

Contribution

This work presents a novel memristor crossbar cryptography architecture combining hypervectors and neural networks for secure data encryption and decryption.

Findings

Achieved 100% decryption accuracy on image and text data.

Demonstrated effective control of memristor non-idealities with increasing hypervector dimensions.

Showed potential of memristor crossbars as unclonable stochastic encoders.

Abstract

We present a novel cryptography architecture based on memristor crossbar array, binary hypervectors, and neural network. Utilizing the stochastic and unclonable nature of memristor crossbar and error tolerance of binary hypervectors and neural network, implementation of the algorithm on memristor crossbar simulation is made possible. We demonstrate that with an increasing dimension of the binary hypervectors, the non-idealities in the memristor circuit can be effectively controlled. At the fine level of controlled crossbar non-ideality, noise from memristor circuit can be used to encrypt data while being sufficiently interpretable by neural network for decryption. We applied our algorithm on image cryptography for proof of concept, and to text en/decryption with 100% decryption accuracy despite crossbar noises. Our work shows the potential and feasibility of using memristor crossbars as an unclonable stochastic encoder unit of cryptography on top of their existing functionality as a vector-matrix multiplication acceleration device.