Synthesis of Adversarial DDOS Attacks Using Tabular Generative Adversarial Networks

TL;DR

This paper explores how Generative Adversarial Networks can create realistic DDoS attack data to test and identify vulnerabilities in Network Intrusion Detection Systems, highlighting potential weaknesses against AI-generated threats.

Contribution

It introduces a method to synthesize adversarial DDoS attacks using GANs and evaluates their impact on IDS, revealing system vulnerabilities to AI-generated threats.

Findings

GAN-synthesized attacks can evade IDS detection

Identifies specific weaknesses in current IDS models

Demonstrates the need for more robust intrusion detection

Abstract

Network Intrusion Detection Systems (NIDS) are tools or software that are widely used to maintain the computer networks and information systems keeping them secure and preventing malicious traffics from penetrating into them, as they flag when somebody is trying to break into the system. Best effort has been set up on these systems, and the results achieved so far are quite satisfying, however, new types of attacks stand out as the technology of attacks keep evolving, one of these attacks are the attacks based on Generative Adversarial Networks (GAN) that can evade machine learning IDS leaving them vulnerable. This project investigates the impact of the Adversarial Attacks synthesized using real DDoS attacks generated using GANs on the IDS. The objective is to discover how will these systems react towards synthesized attacks. marking the vulnerability and weakness points of these systems so we could fix them.