A Real-Time In-Vehicle Network Testbed for Machine Learning-Based IDS Training and Validation

Hesamaldin Jadidbonab; Andrew Tomlinson; Hoang Nga Nguyen; Trang Doan; Siraj Ahmed Shaikh

A Real-Time In-Vehicle Network Testbed for Machine Learning-Based IDS Training and Validation

Hesamaldin Jadidbonab
, Andrew Tomlinson
, Hoang Nga Nguyen
, Trang Doan
, Siraj Ahmed Shaikh

Centre for Future Transport and Cities

Research output: Contribution to journal › Conference article › peer-review

5 Citations (Scopus)

2 Downloads (Pure)

Abstract

Modern vehicles are built of onboard networked computers, known as Electronic Control Units, to realise a range of functionality and features. Due to their connectivity, they are vulnerable to cyberattacks through various vectors including wired (e.g., CAN, Lin) and wireless (e.g., Bluetooth, WiFi) communications. To facilitate building and validating security measures against these attacks, it is vital to replicate the on-board networks and their components, along with a connected external environment representing required use cases and driving scenarios in an instrumental and controllable environment. In this paper, we propose a multi-component testbed representing a flexible and functional in-vehicle architecture for training and validating machine learning-based Intrusion Detection Systems (IDS).

Original language	English
Pages (from-to)	53-68
Number of pages	16
Journal	CEUR Workshop Proceedings
Volume	3125
Publication status	Published - 17 Apr 2022
Event	Workshop on AI and Cybersecurity: co-located with 41st SGAI International Conference on Artificial Intelligence (SGAI 2021) - Cambridge, United Kingdom Duration: 14 Dec 2021 → 14 Dec 2021 https://sites.google.com/view/ai-cybersec-2021/home

Bibliographical note

Publisher Copyright:
© 2021 Copyright for this paper by its authors

Keywords

Automotive security
machine learning
testbed

ASJC Scopus subject areas

General Computer Science

Access to Document

Jadidbonab2022VORFinal published version, 1.14 MBLicence: CC BY

Cite this

@article{76534f74783f411bb0029fbf4059bb4f,

title = "A Real-Time In-Vehicle Network Testbed for Machine Learning-Based IDS Training and Validation",

abstract = "Modern vehicles are built of onboard networked computers, known as Electronic Control Units, to realise a range of functionality and features. Due to their connectivity, they are vulnerable to cyberattacks through various vectors including wired (e.g., CAN, Lin) and wireless (e.g., Bluetooth, WiFi) communications. To facilitate building and validating security measures against these attacks, it is vital to replicate the on-board networks and their components, along with a connected external environment representing required use cases and driving scenarios in an instrumental and controllable environment. In this paper, we propose a multi-component testbed representing a flexible and functional in-vehicle architecture for training and validating machine learning-based Intrusion Detection Systems (IDS).",

keywords = "Automotive security, machine learning, testbed",

author = "Hesamaldin Jadidbonab and Andrew Tomlinson and Nguyen, \{Hoang Nga\} and Trang Doan and Shaikh, \{Siraj Ahmed\}",

note = "Publisher Copyright: {\textcopyright} 2021 Copyright for this paper by its authors; Workshop on AI and Cybersecurity : co-located with 41st SGAI International Conference on Artificial Intelligence (SGAI 2021), AI-Cybersec 2021 ; Conference date: 14-12-2021 Through 14-12-2021",

year = "2022",

month = apr,

day = "17",

language = "English",

volume = "3125",

pages = "53--68",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR-WS",

url = "https://sites.google.com/view/ai-cybersec-2021/home",

}

TY - JOUR

T1 - A Real-Time In-Vehicle Network Testbed for Machine Learning-Based IDS Training and Validation

AU - Jadidbonab, Hesamaldin

AU - Tomlinson, Andrew

AU - Nguyen, Hoang Nga

AU - Doan, Trang

AU - Shaikh, Siraj Ahmed

PY - 2022/4/17

Y1 - 2022/4/17

N2 - Modern vehicles are built of onboard networked computers, known as Electronic Control Units, to realise a range of functionality and features. Due to their connectivity, they are vulnerable to cyberattacks through various vectors including wired (e.g., CAN, Lin) and wireless (e.g., Bluetooth, WiFi) communications. To facilitate building and validating security measures against these attacks, it is vital to replicate the on-board networks and their components, along with a connected external environment representing required use cases and driving scenarios in an instrumental and controllable environment. In this paper, we propose a multi-component testbed representing a flexible and functional in-vehicle architecture for training and validating machine learning-based Intrusion Detection Systems (IDS).

AB - Modern vehicles are built of onboard networked computers, known as Electronic Control Units, to realise a range of functionality and features. Due to their connectivity, they are vulnerable to cyberattacks through various vectors including wired (e.g., CAN, Lin) and wireless (e.g., Bluetooth, WiFi) communications. To facilitate building and validating security measures against these attacks, it is vital to replicate the on-board networks and their components, along with a connected external environment representing required use cases and driving scenarios in an instrumental and controllable environment. In this paper, we propose a multi-component testbed representing a flexible and functional in-vehicle architecture for training and validating machine learning-based Intrusion Detection Systems (IDS).

KW - Automotive security

KW - machine learning

KW - testbed

UR - https://ceur-ws.org/

UR - https://www.scopus.com/pages/publications/85128722799

M3 - Conference article

AN - SCOPUS:85128722799

SN - 1613-0073

VL - 3125

SP - 53

EP - 68

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

T2 - Workshop on AI and Cybersecurity

Y2 - 14 December 2021 through 14 December 2021

ER -

A Real-Time In-Vehicle Network Testbed for Machine Learning-Based IDS Training and Validation

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this