Related Experiment Video
Updated: Jun 29, 2026
03:31
End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
Published on: December 15, 2023
Elevating intrusion detection and security fortification in intelligent networks through cutting-edge machine
Md Minhazul Islam Munna1, Md Mahbubur Rahman1, Jaroslav Frnda2,3
1Department of Computer Science and Technology, Beijing Institute of Technology, 5 Zhongguancun South Street, Beijing, 100081, China.
Scientific Reports
|November 14, 2025
Summary
This study introduces a machine learning framework to enhance Wi-Fi security against KRACK and Kr00k attacks. The novel ensemble model significantly improves intrusion detection accuracy for IoT devices.
Area of Science:
- Cybersecurity
- Machine Learning
- Network Security
Background:
- Internet of Things (IoT) devices rely heavily on Wi-Fi, creating vulnerabilities exploited by attacks like KRACK and Kr00k.
- Traditional Intrusion Detection Systems (IDS) struggle with model overfitting, feature extraction, and high false positive rates.
Purpose of the Study:
- To develop a robust multiclass machine learning intrusion detection framework for Wi-Fi networks.
- To address the limitations of traditional IDS in detecting sophisticated attacks on IoT devices.
Main Methods:
- Implemented advanced feature selection techniques for critical attribute identification.
- Developed two machine learning architectures: a baseline classifier and a stacked ensemble model.
- The ensemble model incorporated noise injection, Principal Component Analysis (PCA), and meta-learning.
Main Results:
- The proposed ensemble architecture achieved 98% accuracy, 98% precision, and 98% recall.
- The ensemble model demonstrated a low false positive rate of only 2%.
- Performance surpassed existing state-of-the-art methods on the AWID3 dataset.
Conclusions:
- Combining preprocessing strategies with ensemble learning effectively fortifies network security against Wi-Fi attacks.
- The developed framework offers a scalable and reliable solution for securing IoT environments.
- Future work includes real-time deployment and adversarial resilience testing.
