A Review of Detecting DDoS Attacks Based on Entropy Computation

• Published in: International Conference on System Modeling & Advancement in Research Trends (Online) pp. 146 - 153
• Main Authors: Agrawal, Atul, Baniya, Pashupati, Gupta, Bishnu Bahadur, Chaturvedi, Saumya, Singh, Gaurav Kumar, Yadav, Deepak
• Format: Conference Proceeding
• Language: English
• Published: IEEE 22.12.2023
• Subjects: Artificial intelligence; Attack type; Behavioral sciences; Business; Computer crime; DDoS Attack; Deep learning; deep learning-based detection techniques; Denial-of-service attack; Detection; Entropy; Statistical techniques; Web and internet services
• ISBN: 9798350369861
• ISSN: 2767-7362
• DOI: 10.1109/SMART59791.2023.10428656

The current research examines the global issue of distributed denial of service (DDoS) attacks and their disruptive effects on internet services around the globe. It focuses on the financial and operational consequences these attacks impose on different businesses and entities. The objective is to demonstrate how deep learning models can effectively identify both DoS (Denial of Service) and DDoS attacks by leveraging historical and contemporary datasets. By highlighting the combination of innovative techniques and advanced datasets, the paper presents a promising approach to enhance cyber security and strengthen networks against evolving threats. This paper's thorough examination of recent research endeavors led to the conclusion that several cutting-edge technologies, including artificial intelligence, deep learning, fuzzy logic, and others, may be employed in the future to lessen DDoS and DoS attacks. Additionally, future papers will be introducing a novel, speedier entropy-based technique for detecting these attacks through flow-based analysis. The overall goal of the research is to advance the subject of cyber security and offer suggestions for lessening the effects of DDoS attacks.