Abstract: A Distributed Denial-Of-Service (DDoS) assault is a hostile activity that aims to overload a server, service, or network with excessive network traffic, rendering it unusable. The fact that attack patterns are constantly shifting, cyber offense techniques are developing quickly, and cyber offense materials are freely available on the dark web have made this a difficult task. DDoS assaults have the potential to seriously impair online services, resulting in lost profits, harmed reputations, and dwindling client confidence. Additionally, it may lead to overheating or power outages, which could harm infrastructure. Taking into account all of these factors, this research focuses on predicting and classifying DDoS attacks through the analysis of network traffic data using a deep learning technique to automate the manual process. In the end, this can reduce human error in the detection process and save time and effort. This project trains the "DDoS Evaluation – CICDDoS2019" dataset using the power of the Long Short Term Memory(LSTM). To improve the accuracy of anomaly detection, the LSTM is trained in an unsupervised environment to recover encoded data. In a monitored setting, the LSTM is trained to categorize network traffic data into DDoS attacks.
Keywords: Distributed Denial-Of-Service (DDoS), Deep Learning, Long Short Term Memory(LSTM)
Cite:
Sharan K R, Shreekavya C H, Rony Dominic, Soniya A Gunagi, Vasudha G Rao, "Detection of DDoS Attack Using Deep Learning", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.13343.
|
DOI:
10.17148/IJARCCE.2024.13343
