A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack

In recent years, many side-channel attack (SCA) based on deep learning have emerged, making it possible to break protected encryption algorithms. How- ever, since the training of deep learning is based on back-propagation, a long training time is required. In deep learning SCA, because of the encryp...

Full description

Saved in:
Bibliographic Details
Main Author: Huang, Xuyang
Other Authors: Goh Wang Ling
Format: Thesis-Master by Coursework
Language:English
Published: Nanyang Technological University 2022
Subjects:
Online Access:https://hdl.handle.net/10356/154903
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-154903
record_format dspace
spelling sg-ntu-dr.10356-1549032023-07-04T15:19:12Z A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack Huang, Xuyang Goh Wang Ling School of Electrical and Electronic Engineering Institute of Microelectronics (IME), A*STAR Singapore EWLGOH@ntu.edu.sg Engineering::Electrical and electronic engineering::Microelectronics In recent years, many side-channel attack (SCA) based on deep learning have emerged, making it possible to break protected encryption algorithms. How- ever, since the training of deep learning is based on back-propagation, a long training time is required. In deep learning SCA, because of the encryption algorithm, it is common to train multiple models based on the number of subkeys in one time attack, so the training time is multiplied as a drawback of DL- SCA. This work presented new Deep learning Side-channel Attack (DL-SCA) models that are based on Extreme Learning Machine (ELM). Unlike the conventional iterative backpropagation method, ELM is a fast learning algorithm that computes the trainable weights within a single iteration. Two models (Ensemble bpELM and CAE-ebpELM) are designed to perform SCA on AES with Boolean masking and desynchronization/jittering. The best models for both at- tack tasks can be trained 27× faster than MLP and 5× faster than CNN respectively. Verified and validated using ASCAD dataset, our models successfully recover all 16 subkeys using approximately 3K traces in the worst case scenario. Master of Science (Electronics) 2022-01-14T03:08:20Z 2022-01-14T03:08:20Z 2021 Thesis-Master by Coursework Huang, X. (2021). A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/154903 https://hdl.handle.net/10356/154903 en application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Electrical and electronic engineering::Microelectronics
spellingShingle Engineering::Electrical and electronic engineering::Microelectronics
Huang, Xuyang
A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack
description In recent years, many side-channel attack (SCA) based on deep learning have emerged, making it possible to break protected encryption algorithms. How- ever, since the training of deep learning is based on back-propagation, a long training time is required. In deep learning SCA, because of the encryption algorithm, it is common to train multiple models based on the number of subkeys in one time attack, so the training time is multiplied as a drawback of DL- SCA. This work presented new Deep learning Side-channel Attack (DL-SCA) models that are based on Extreme Learning Machine (ELM). Unlike the conventional iterative backpropagation method, ELM is a fast learning algorithm that computes the trainable weights within a single iteration. Two models (Ensemble bpELM and CAE-ebpELM) are designed to perform SCA on AES with Boolean masking and desynchronization/jittering. The best models for both at- tack tasks can be trained 27× faster than MLP and 5× faster than CNN respectively. Verified and validated using ASCAD dataset, our models successfully recover all 16 subkeys using approximately 3K traces in the worst case scenario.
author2 Goh Wang Ling
author_facet Goh Wang Ling
Huang, Xuyang
format Thesis-Master by Coursework
author Huang, Xuyang
author_sort Huang, Xuyang
title A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack
title_short A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack
title_full A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack
title_fullStr A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack
title_full_unstemmed A backpropagation extreme learning machine approach to fast training neural network-based side-channel attack
title_sort backpropagation extreme learning machine approach to fast training neural network-based side-channel attack
publisher Nanyang Technological University
publishDate 2022
url https://hdl.handle.net/10356/154903
_version_ 1772825471177719808