An effective approach of detecting DDoS using Artificial Neural Networks

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

23 Scopus citations

Abstract

One of the major problems with internet is its security and DDoS is one of the main and devastating security problems that decline the network and resource performance by sending simultaneous request traffic to the target network or system. The attackers even use the legitimate traffic and sometimes even change the traffic information to pass through the detection systems. In this research article, a novel DDoS detection mechanism is proposed based on Artificial Neural Networks principles. This method analyzes the system resources and the network data in order to train the ANN DDoS detection system to detect normal and abnormal traffic. The legitimate traffic is allowed to pass through and abnormal traffic is flagged suspicious to go through detection and defense system. This ANN based DDoS detection method provided good results in detecting DDoS attacks.

Original languageEnglish
Title of host publicationProceedings of the 2017 International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages707-711
Number of pages5
ISBN (Electronic)9781509044412
DOIs
StatePublished - 2 Jul 2017
Event2nd IEEE International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2017 - Chennai, India
Duration: 22 Mar 201724 Mar 2017

Publication series

NameProceedings of the 2017 International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2017
Volume2018-January

Conference

Conference2nd IEEE International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2017
Country/TerritoryIndia
CityChennai
Period22/03/1724/03/17

Keywords

  • ANN
  • DDoS
  • Security
  • Smurf

Fingerprint

Dive into the research topics of 'An effective approach of detecting DDoS using Artificial Neural Networks'. Together they form a unique fingerprint.

Cite this