A Distributed Denial-of-Service (DDoS) Attack Detection Approach in Fog Layer Based on Distributed Blockchain Database and Machine Learning

Message:
Article Type:
Research/Original Article (دارای رتبه معتبر)
Abstract:

DDoS attacks make network services unavailable to users by sending fake traffic by botnets. One of the methods to deal with DDoS attacks is to use machine learning, but these methods face challenges such as high volume of IoT traffic and data imbalance. This paper introduces a distributed intrusion detection system in the fog layer that detects network attack traffic in a decentralized manner. In this method, each fog node acts as an intrusion detection system, and by exchanging blacklists through the blockchain, they increase the secrecy of detecting attacks. Fog nodes identify the main features of network traffic using the Coati optimization algorithm and use these features to train a multilayer neural network in intrusion detection. The selection of features reduces traffic and increases the accuracy and speed of attack detection. Based on game theory, the GAN method is used to balance network traffic. Tests performed in the MATLAB and on the NSL-KDD show that the proposed system has accuracy, sensitivity, and precision of 98.67%, 98.52%, and 98.34%, respectively. This method is more accurate in identifying network attacks than feature selection methods such as WOA, GWO, and HHO and more accurate than LSTM and CNN.

Language:
Persian
Published:
Journal of Southern Communication Engineering, Volume:14 Issue: 53, 2024
Pages:
67 to 90
https://www.magiran.com/p2772239  
سامانه نویسندگان
  • Eghbali، Mohsen
    Author (1)
    Eghbali, Mohsen
    Phd Student Computer Engineering, Faculty of Technology and Engineering, Maybod Branch, Islamic Azad University, Meybod, Iran
اطلاعات نویسنده(گان) توسط ایشان ثبت و تکمیل شده‌است. برای مشاهده مشخصات و فهرست همه مطالب، صفحه رزومه را ببینید.
مقالات دیگری از این نویسنده (گان)