Citation

Abuajwa, Osama and Mohd Hassan, Siti Maisurah and Mahmud, Azwan and Abdul Aziz, Azlan (2025) Deep Packet Inspection (DPI) Technologies and Their Role in Cyber Threat Detection. In: 2025 Multimedia University Engineering Conference, MECON 2025, 21 July 2025 - 23 July 2025, Cyberjaya, Malaysia.

Text 49.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Deep Packet Inspection (DPI), which looks at both packet headers and payloads, has emerged as a crucial technique for preventing such threats and enabling the realtime identification of anomalies and malicious behaviour. However, DPI’s processing challenges and high computing costs limit its use in high-speed networks. Furthermore, due to evolving attack techniques, traditional Malware Traffic Classification (MTC) methodologies, like port-based and DPIbased approaches, are no longer as effective. In order to combat threats like malware, advanced persistent threats (APTs), and distributed denial-of-service (DDoS) attacks, strong detection is now more critical than ever due to the quick rise in network traffic. DPI and Intrusion Detection and Prevention Systems (IDS/IPS) are essential tools for examining packet payloads and headers across Transmission Control Protocol/Internet Protocol (TCP/IP) layers for real-time monitoring and anomaly detection. This work thoroughly analyses DPI for network security, emphasising machine learning-based methods, algorithms, and detection strategies. The review finds ways to improve DPI-based security solutions by assessing these techniques' accuracy, performance, and computational efficiency. Nevertheless, the review reveals that preserving accuracy with by machine learning (ML) algorithms such as ML-Intelligent Botnet Detection system integrated-DPI (ML-IBotD-DPI) may result in higher false positive rates, as demonstrated by ML-Hybrid-DPI's . The findings address the need for all metrics to be balanced in real-world applications, concerns about known and new threats, and how to create scala

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Machine Learning
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Engineering and Technology (FET) Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	18 Mar 2026 04:33
Last Modified:	18 Mar 2026 05:30
URII:	http://shdl.mmu.edu.my/id/eprint/15539

Downloads

Edit (login required)