The Distributed Denial of Service (DDoS) attack is one of the most dangerous cyberattacks on the Internet, so can affect any server on any type of network, causing connectivity problems and even total loss of services. Machine learning can solve computational security problems and is frequently used to defend against cyber attacks. This article proposes the construction of a network topology where several DDoS attacks were applied, which will be detected by three Machine Learning classification algorithms. A dataset was generated from the collection of packets circulating in the network with samples of normal traffic and malicious packets, on which the experimental tests were carried out. In the classification task, the best performing supervised learning algorithm was Random Forest, with an accuracy of 100%. Finally, upon detecting a DDoS attack on the network, Dijkstra's optimization algorithm is applied to find an alternative route to mitigate network oversaturation. Two scenarios were proposed, the first analyzes the optimal route in an attacked network and the second without attacks. The results show a reconfiguration in the network to avoid routes where DDoS attack detection was applied.
|Title of host publication||2022 IEEE 2nd International Conference on Computer Communication and Artificial Intelligence, CCAI 2022|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|Number of pages||7|
|State||Published - 2022|
|Event||2nd IEEE International Conference on Computer Communication and Artificial Intelligence, CCAI 2022 - Beijing, China|
Duration: 6 May 2022 → 8 May 2022
|Name||2022 IEEE 2nd International Conference on Computer Communication and Artificial Intelligence, CCAI 2022|
|Conference||2nd IEEE International Conference on Computer Communication and Artificial Intelligence, CCAI 2022|
|Period||6/05/22 → 8/05/22|
Bibliographical noteFunding Information:
This work was supported by IDEIAGEOCA Research Group of Universidad Politécnica Salesiana in Quito, Ecuador.
© 2022 IEEE.
- Dijkstra algorithm
- logistic regression
- machine learning
- malicious traffic detection
- random forest
- support vector machine