Publication

Abstract :

The Internet is now a crucial aspect of our lives, and distributed computing has grown significantly in size, capability, and complexity, becoming an essential element of our daily lives. There are numerous vulnerabilities in distributed network computing, including viruses, worms, distributed denial of service (DDoS), DoS, and more. One of the more newsworthy attacks recently has been distributed denial-of-service attacks. Attacks that cause denials of service (DoS) have developed and had an influence on the Internet infrastructure. A DDoS attack aims to prevent the targeted users from accessing a computer resource. In this paper, we perform a comprehensive survey of detecting DDoS packets via couple of machine learning algorithms such as Random Forest, Gradient Descent, decision tree, Multi-layer Perceptron, and radial basis function using the CIC-DDoS2019 dataset. We also combine all our machine learning models into one hybrid model via an ensemble classifier. After that, we experiment with a DDoS attack on a local server via Ufonet and capture the data packets during the attack using the Wireshark tool. The captured data will be converted to a CSV file and can be given as input to machine learning models. The results show that each machine learning algorithm is predicting DDoS packets incoming into the server. The experimental results show that the proposed DDoS attack detection method based on machine learning has a good detection rate for the current popular DDoS attack. Finally, directions for future research work have been pointed out.