Side-Channel Analysis of Integrate-and-Fire Neurons within Spiking Neural Networks

Matthias Probst; Manuel Brosch; Georg Sigl

Paper 2023/505

Side-Channel Analysis of Integrate-and-Fire Neurons within Spiking Neural Networks

Matthias Probst

, Technical University of Munich

Manuel Brosch

, Technical University of Munich

Georg Sigl

, Technical University of Munich, Fraunhofer Institute for Applied and Integrated Security

Abstract

Spiking neural networks gain attention due to low power properties and event-based operation, making them suitable for usage in resource constrained embedded devices. Such edge devices allow physical access opening the door for side-channel analysis. In this work, we reverse engineer the parameters of a feed-forward spiking neural network implementation with correlation power analysis. Localized measurements of electro-magnetic emanations enable our attack, despite inherent parallelism and the resulting algorithmic noise of the network. We provide a methodology to extract valuable parameters of integrate-and-fire neurons in all layers, as well as the layer sizes.

Metadata

Available format(s): PDF
Category: Attacks and cryptanalysis
Publication info: Preprint.
Keywords: Side-Channel Analysis Neural Networks Spiking Neural Networks
Contact author(s): matthias probst @ tum de
manuel brosch @ tum de
sigl @ tum de
History: 2023-04-11: revised; 2023-04-07: received; See all versions
Short URL: https://ia.cr/2023/505
License: CC BY

BibTeX

@misc{cryptoeprint:2023/505,
      author = {Matthias Probst and Manuel Brosch and Georg Sigl},
      title = {Side-Channel Analysis of Integrate-and-Fire Neurons within Spiking Neural Networks},
      howpublished = {Cryptology {ePrint} Archive, Paper 2023/505},
      year = {2023},
      url = {https://eprint.iacr.org/2023/505}
}