Towards realistic problem-space adversarial attacks against machine learning in network intrusion detection

被引:0
|
作者
Catillo, Marta [1 ]
Pecchia, Antonio [1 ]
Repola, Antonio [1 ]
Villano, Umberto [1 ]
机构
[1] Univ Sannio, Benevento, Italy
来源
19TH INTERNATIONAL CONFERENCE ON AVAILABILITY, RELIABILITY, AND SECURITY, ARES 2024 | 2024年
关键词
intrusion detection; machine learning; adversarial examples; supervised learning; Denial of Service; DETECTION SYSTEMS; ROBUSTNESS;
D O I
10.1145/3664476.3669974
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Current trends in network intrusion detection systems (NIDS) capitalize on the extraction of features from network traffic and the use of up-to-date machine and deep learning techniques to infer a detection model; in consequence, NIDS can be vulnerable to adversarial attacks. Differently from the plethora of contributions that apply (and misuse) feature-level attacks envisioned in application domains far from NIDS, this paper proposes a novel approach to adversarial attacks, which consists in a realistic problem-space perturbation of the network traffic. The perturbation is achieved through a traffic control utility. Experiments are based on normal and Denial of Service traffic in both legitimate and adversarial conditions, and the application of four popular techniques to learn the NIDS models. The results highlight the transferability of the adversarial examples generated by the proposed problem-space attack as well as the effectiveness at inducing traffic misclassifications across the NIDS models obtained.
引用
收藏
页数:8
相关论文
共 50 条
  • [31] Using Machine Learning Algorithms and Honeypot Systems to Detect Adversarial Attacks on Intrusion Detection Systems
    P. E. Yugai
    D. A. Moskvin
    Automatic Control and Computer Sciences, 2024, 58 (8) : 1226 - 1233
  • [32] Adversarial Examples Against the Deep Learning Based Network Intrusion Detection Systems
    Yang, Kaichen
    Liu, Jianqing
    Zhang, Chi
    Fang, Yuguang
    2018 IEEE MILITARY COMMUNICATIONS CONFERENCE (MILCOM 2018), 2018, : 559 - 564
  • [33] Addressing Adversarial Attacks Against Security Systems Based on Machine Learning
    Apruzzese, Giovanni
    Colajanni, Michele
    Ferretti, Luca
    Marchetti, Mirco
    2019 11TH INTERNATIONAL CONFERENCE ON CYBER CONFLICT (CYCON): SILENT BATTLE, 2019, : 383 - 400
  • [34] Towards adversarial realism and robust learning for IoT intrusion detection and classification
    Vitorino, Joao
    Praca, Isabel
    Maia, Eva
    ANNALS OF TELECOMMUNICATIONS, 2023, 78 (7-8) : 401 - 412
  • [35] Towards adversarial realism and robust learning for IoT intrusion detection and classification
    João Vitorino
    Isabel Praça
    Eva Maia
    Annals of Telecommunications, 2023, 78 : 401 - 412
  • [36] Anomaly-Based Intrusion Detection by Machine Learning: A Case Study on Probing Attacks to an Institutional Network
    Tufan, Emrah
    Tezcan, Cihangir
    Acarturk, Cengiz
    IEEE ACCESS, 2021, 9 : 50078 - 50092
  • [37] Network intrusion detection system: A machine learning approach
    Panda, Mrutyunjaya
    Abraham, Ajith
    Das, Swagatam
    Patra, Manas Ranjan
    INTELLIGENT DECISION TECHNOLOGIES-NETHERLANDS, 2011, 5 (04): : 347 - 356
  • [38] Adversarial Attacks on Network Intrusion Detection Systems Using Flow Containers
    Liu, Tzong-Jye
    COMPUTER JOURNAL, 2024, 67 (02): : 728 - 745
  • [39] A Framework for Generating Evasion Attacks for Machine Learning Based Network Intrusion Detection Systems
    Mogg, Raymond
    Enoch, Simon Yusuf
    Kim, Dong Seong
    INFORMATION SECURITY APPLICATIONS, 2021, 13009 : 51 - 63
  • [40] A machine learning-based intrusion detection for detecting internet of things network attacks
    Saheed, Yakub Kayode
    Abiodun, Aremu Idris
    Misra, Sanjay
    Holone, Monica Kristiansen
    Colomo-Palacios, Ricardo
    ALEXANDRIA ENGINEERING JOURNAL, 2022, 61 (12) : 9395 - 9409