Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Robust Intrusion Detection System with Explainable Artificial Intelligence

Betül Güvenç Paltun, Ramin Fuladi, Rim El Malki

TL;DR

This work tackles adversarial vulnerabilities in ML-based intrusion detection within O-RAN by proposing an agnostic, XAI-driven framework that uses SHAP value distributions to detect adversarial perturbations in real time and implements zero-touch mitigation. The approach is validated in a real OpenAirInterface-based RRC signaling storm scenario, combining an autoencoder-based IDS with SHAP/LIME/Permutation explanations, where SHAP demonstrates superior accuracy ($0.9259$) and F1 ($0.8654$) and recall of 1.0. The results show the method’s ability to identify manipulated inputs and respond promptly, enhancing IDS robustness in multi-vendor, open-interface network environments. Overall, the paper contributes a practical, explainable, and autonomous defense mechanism for protecting critical 5G/6G network operations against sophisticated adversarial threats.

Abstract

Machine learning (ML) models serve as powerful tools for threat detection and mitigation; however, they also introduce potential new risks. Adversarial input can exploit these models through standard interfaces, thus creating new attack pathways that threaten critical network operations. As ML advancements progress, adversarial strategies become more advanced, and conventional defenses such as adversarial training are costly in computational terms and often fail to provide real-time detection. These methods typically require a balance between robustness and model performance, which presents challenges for applications that demand instant response. To further investigate this vulnerability, we suggest a novel strategy for detecting and mitigating adversarial attacks using eXplainable Artificial Intelligence (XAI). This approach is evaluated in real time within intrusion detection systems (IDS), leading to the development of a zero-touch mitigation strategy. Additionally, we explore various scenarios in the Radio Resource Control (RRC) layer within the Open Radio Access Network (O-RAN) framework, emphasizing the critical need for enhanced mitigation techniques to strengthen IDS defenses against advanced threats and implement a zero-touch mitigation solution. Extensive testing across different scenarios in the RRC layer of the O-RAN infrastructure validates the ability of the framework to detect and counteract integrated RRC-layer attacks when paired with adversarial strategies, emphasizing the essential need for robust defensive mechanisms to strengthen IDS against complex threats.

Robust Intrusion Detection System with Explainable Artificial Intelligence

TL;DR

This work tackles adversarial vulnerabilities in ML-based intrusion detection within O-RAN by proposing an agnostic, XAI-driven framework that uses SHAP value distributions to detect adversarial perturbations in real time and implements zero-touch mitigation. The approach is validated in a real OpenAirInterface-based RRC signaling storm scenario, combining an autoencoder-based IDS with SHAP/LIME/Permutation explanations, where SHAP demonstrates superior accuracy () and F1 () and recall of 1.0. The results show the method’s ability to identify manipulated inputs and respond promptly, enhancing IDS robustness in multi-vendor, open-interface network environments. Overall, the paper contributes a practical, explainable, and autonomous defense mechanism for protecting critical 5G/6G network operations against sophisticated adversarial threats.

Abstract

Machine learning (ML) models serve as powerful tools for threat detection and mitigation; however, they also introduce potential new risks. Adversarial input can exploit these models through standard interfaces, thus creating new attack pathways that threaten critical network operations. As ML advancements progress, adversarial strategies become more advanced, and conventional defenses such as adversarial training are costly in computational terms and often fail to provide real-time detection. These methods typically require a balance between robustness and model performance, which presents challenges for applications that demand instant response. To further investigate this vulnerability, we suggest a novel strategy for detecting and mitigating adversarial attacks using eXplainable Artificial Intelligence (XAI). This approach is evaluated in real time within intrusion detection systems (IDS), leading to the development of a zero-touch mitigation strategy. Additionally, we explore various scenarios in the Radio Resource Control (RRC) layer within the Open Radio Access Network (O-RAN) framework, emphasizing the critical need for enhanced mitigation techniques to strengthen IDS defenses against advanced threats and implement a zero-touch mitigation solution. Extensive testing across different scenarios in the RRC layer of the O-RAN infrastructure validates the ability of the framework to detect and counteract integrated RRC-layer attacks when paired with adversarial strategies, emphasizing the essential need for robust defensive mechanisms to strengthen IDS against complex threats.

Paper Structure

This paper contains 17 sections, 2 equations, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. Adversarial AI
  4. Explainable AI
  5. RRC Signaling Storm Attacks
  6. Related Work
  7. Proposed Framework
  8. Adversarial attacks against IDS
  9. Detection of Adversarial attacks
  10. Experimental Evaluation
  11. O-RAN Setup
  12. Experiment Settings
  13. Data
  14. Intrusion Detection System
  15. Scenarios
  16. ...and 2 more sections

Figures (5)

  • Figure 1: General structure of the problem and proposed XAI-based adversarial attack detection framework.
  • Figure 2: Steps of xApp feature retrieval from a gNB within OAI setup using FlexRIC.
  • Figure 3: The key features used for the detection of RRC signaling storms.
  • Figure 4: The accuracy of IDS under different adversarial attack methods.
  • Figure 5: Method comparison with XAI-based mitigation