Download PDFOpen PDF in browser

DDoS Attack Detection on SDN with Conv1D and LSTM

EasyChair Preprint no. 8573

5 pagesDate: August 3, 2022


Especially in the last decade, many companies have been digitalizing their architectures and using various kinds of software and hardware that reside in databases or virtual hosts within Local Area Networks. With the latest dynamic technologies, network data flow is monitored and controlled swiftly and in detail with SDN (Software Defined Networks). SDN packet traffic has a structure that can be easily projected compared to traditional networks. It provides broader control possibilities on the network and can be controlled faster. In this research, an SDN is used for identifying or increasing the quality of the identification of various network data and summarizes for further investigation. It is easier to record data flow from networks, analyze the network, and detect various types of malware, DoS, and DDoS (Distributed Denial of Services); SDN software is used to categorize network data for security and higher performance. Deep Learning methods have been very efficient for classifying different types by their features. In this proposed model with Conv1d (One Dimensional Convolutional Neural Network), LSTM (Long Short Term Memory), and ANN (Artificial Neural Networks), the performance results were found to be satisfactory to categorize malware or DDoS within healthy dataflow. It is planned to use the information for future work in the industry and further research into detecting malware, DoS, and DDoS with higher performance.

Keyphrases: attack detection, Convolutional Neural Networks, DDoS, deep learning, Software Defined Network

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
  author = {Alperen Örsdemir and Hamidullah Nazari and Devrim Akgün},
  title = {DDoS Attack Detection on SDN with Conv1D and LSTM},
  howpublished = {EasyChair Preprint no. 8573},

  year = {EasyChair, 2022}}
Download PDFOpen PDF in browser