Attack rules: an adversarial approach to generate attacks for Industrial Control Systems using machine learning

Muhammad Azmi Umer; Chuadhry Mujeeb Ahmed; Muhammad Taha Jilani; Aditya P. Mathur

doi:10.1145/3462633.3483976

Attack rules: an adversarial approach to generate attacks for Industrial Control Systems using machine learning

Muhammad Azmi Umer, Chuadhry Mujeeb Ahmed, Muhammad Taha Jilani, Aditya P. Mathur

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

9 Citations (Scopus)

23 Downloads (Pure)

Abstract

Adversarial learning is used to test the robustness of machine learning algorithms under attack and create attacks that deceive the anomaly detection methods in Industrial Control System (ICS). Given that security assessment of an ICS demands that an exhaustive set of possible attack patterns is studied, in this work, we propose an association rule mining-based attack generation technique. The technique has been implemented using data from a Secure Water Treatment plant. The proposed technique was able to generate more than 110,000 attack patterns constituting a vast majority of new attack vectors which were not seen before. Automatically generated attacks improve our understanding of the potential attacks and enable the design of robust attack detection techniques.

Original language	English
Title of host publication	CPSIoTSec 2021 - Proceedings of the 2nd Workshop on CPS and IoT Security and Privacy, co-located with CCS 2021
Place of Publication	New York, NY.
Pages	35–40
Number of pages	6
DOIs	https://doi.org/10.1145/3462633.3483976
Publication status	Published - 15 Nov 2021
Event	CPSIoTSec 2021: The 2nd Joint Workshop on CPS & IoT Security and Privacy - Seoul, Korea, Republic of Duration: 15 Nov 2021 → 15 Nov 2021 https://cpsiotsec.github.io/

Conference

Conference	CPSIoTSec 2021
Country/Territory	Korea, Republic of
City	Seoul
Period	15/11/21 → 15/11/21
Internet address	https://cpsiotsec.github.io/

Keywords

attack detection
attack generation
ICS security
machine learning
adversarial learning
association rule mining

Access to Document

10.1145/3462633.3483976

Umer-etal-CPSIoTSec-2021-an-adversarial-approach-to-generate-attacks-for-Industrial-Control-SystemsAccepted author manuscript, 4.28 MB

Cite this

@inproceedings{45189c9edfff4ae38588f083d349b773,

title = "Attack rules: an adversarial approach to generate attacks for Industrial Control Systems using machine learning",

abstract = "Adversarial learning is used to test the robustness of machine learning algorithms under attack and create attacks that deceive the anomaly detection methods in Industrial Control System (ICS). Given that security assessment of an ICS demands that an exhaustive set of possible attack patterns is studied, in this work, we propose an association rule mining-based attack generation technique. The technique has been implemented using data from a Secure Water Treatment plant. The proposed technique was able to generate more than 110,000 attack patterns constituting a vast majority of new attack vectors which were not seen before. Automatically generated attacks improve our understanding of the potential attacks and enable the design of robust attack detection techniques.",

keywords = "attack detection, attack generation, ICS security, machine learning, adversarial learning, association rule mining",

author = "Umer, {Muhammad Azmi} and Ahmed, {Chuadhry Mujeeb} and Jilani, {Muhammad Taha} and Mathur, {Aditya P.}",

year = "2021",

month = nov,

day = "15",

doi = "10.1145/3462633.3483976",

language = "English",

isbn = "9781450384872",

pages = "35–40",

booktitle = "CPSIoTSec 2021 - Proceedings of the 2nd Workshop on CPS and IoT Security and Privacy, co-located with CCS 2021",

note = "CPSIoTSec 2021 : The 2nd Joint Workshop on CPS & IoT Security and Privacy ; Conference date: 15-11-2021 Through 15-11-2021",

url = "https://cpsiotsec.github.io/",

}

Umer, MA, Ahmed, CM, Jilani, MT & Mathur, AP 2021, Attack rules: an adversarial approach to generate attacks for Industrial Control Systems using machine learning. in CPSIoTSec 2021 - Proceedings of the 2nd Workshop on CPS and IoT Security and Privacy, co-located with CCS 2021. New York, NY., pp. 35–40, CPSIoTSec 2021, Seoul, Korea, Republic of, 15/11/21. https://doi.org/10.1145/3462633.3483976

Attack rules: an adversarial approach to generate attacks for Industrial Control Systems using machine learning. / Umer, Muhammad Azmi; Ahmed, Chuadhry Mujeeb; Jilani, Muhammad Taha et al.
CPSIoTSec 2021 - Proceedings of the 2nd Workshop on CPS and IoT Security and Privacy, co-located with CCS 2021. New York, NY., 2021. p. 35–40.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

TY - GEN

T1 - Attack rules

T2 - CPSIoTSec 2021

AU - Umer, Muhammad Azmi

AU - Ahmed, Chuadhry Mujeeb

AU - Jilani, Muhammad Taha

AU - Mathur, Aditya P.

PY - 2021/11/15

Y1 - 2021/11/15

N2 - Adversarial learning is used to test the robustness of machine learning algorithms under attack and create attacks that deceive the anomaly detection methods in Industrial Control System (ICS). Given that security assessment of an ICS demands that an exhaustive set of possible attack patterns is studied, in this work, we propose an association rule mining-based attack generation technique. The technique has been implemented using data from a Secure Water Treatment plant. The proposed technique was able to generate more than 110,000 attack patterns constituting a vast majority of new attack vectors which were not seen before. Automatically generated attacks improve our understanding of the potential attacks and enable the design of robust attack detection techniques.

AB - Adversarial learning is used to test the robustness of machine learning algorithms under attack and create attacks that deceive the anomaly detection methods in Industrial Control System (ICS). Given that security assessment of an ICS demands that an exhaustive set of possible attack patterns is studied, in this work, we propose an association rule mining-based attack generation technique. The technique has been implemented using data from a Secure Water Treatment plant. The proposed technique was able to generate more than 110,000 attack patterns constituting a vast majority of new attack vectors which were not seen before. Automatically generated attacks improve our understanding of the potential attacks and enable the design of robust attack detection techniques.

KW - attack detection

KW - attack generation

KW - ICS security

KW - machine learning

KW - adversarial learning

KW - association rule mining

U2 - 10.1145/3462633.3483976

DO - 10.1145/3462633.3483976

M3 - Conference contribution book

SN - 9781450384872

SP - 35

EP - 40

BT - CPSIoTSec 2021 - Proceedings of the 2nd Workshop on CPS and IoT Security and Privacy, co-located with CCS 2021

CY - New York, NY.

Y2 - 15 November 2021 through 15 November 2021

ER -

Attack rules: an adversarial approach to generate attacks for Industrial Control Systems using machine learning

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this