Advanced Detection of Application Layer DDoS Attacks Using Signature Algorithms

Application-layer Distributed Denial of Service (App-DDoS) attacks continue to pose significant challenges in cybersecurity. These attacks exploit the flexibility of HTTP request headers, allowing attackers to craft requests that mimic legitimate traffic. Recent studies highlight the lack of current datasets and research on forged request headers, which complicates detection efforts. A new detection method utilizing advanced signature detection algorithms has been developed, achieving a high accuracy of 96.93% in identifying malicious traffic. This method effectively categorizes traffic before it reaches the web server, addressing a critical gap in existing defenses. The research emphasizes the importance of using recent datasets to improve detection capabilities. The findings have implications for real-world applications, suggesting that signature-based detection remains a viable option alongside machine learning approaches. The study was supported by Universiti Teknologi Malaysia and involved a practical analysis of attack strategies.

Key Points: • New signature detection methods achieve 96.93% accuracy in identifying App-DDoS attacks. • Research highlights the challenges of detecting forged request headers in HTTP DDoS attacks. • Recent datasets are crucial for improving detection capabilities against evolving attack patterns.