Density peak-based clustering of industrial control protocols for reverse engineering

Di Tong; Yongjun Wang

doi:10.1117/12.2635863

5 May 2022 Density peak-based clustering of industrial control protocols for reverse engineering

Di Tong, Yongjun Wang

Proceedings Volume 12245, International Conference on Cryptography, Network Security, and Communication Technology (CNSCT 2022); 122450C (2022) https://doi.org/10.1117/12.2635863
Event: International Conference on Cryptography, Network Security, and Communication Technology (CNSCT 2022), 2022, Sanya, China

Abstract

Message clustering is the first and most crucial step in protocol reverse engineering, and the accuracy of protocol clustering directly affects protocol format extraction and state machine inference. The existing text protocol clustering methods do not apply to industrial control protocols, and the number of clusters and cluster centers cannot be determined automatically. The n-gram model is used to extract the protocol message features and construct the message feature vector for the clustering algorithm, which preserves the feature information in the protocol messages; density peak clustering is used to avoid the problem that the initial number of clusters and cluster centers cannot be determined automatically, and the automatic clustering of industrial control protocols is realized. It is tested on three commonly used industrial control protocol datasets, Modbus, DNP3, and S7common, and achieves good results on two metrics, purity and F1 score.

Citation Download Citation

Di Tong and Yongjun Wang "Density peak-based clustering of industrial control protocols for reverse engineering", Proc. SPIE 12245, International Conference on Cryptography, Network Security, and Communication Technology (CNSCT 2022), 122450C (5 May 2022); https://doi.org/10.1117/12.2635863

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available