Ensemble Model to Detect Wireless Attacks in Mobile Ad hoc Networks
N. Ravi1, G. Ramachandran2
1N. Ravi, Research Scholar, Department of Computer and Information Sciences, Annamalai University, Annamalainagar (Tamil Nadu), India.
2Dr. G. Ramachandran, Assistant Professor, Department of Computer and Information Sciences, Annamalai University, Annamalainagar (Tamil Nadu), India.
Manuscript received on 25 May 2019 | Revised Manuscript received on 03 June 2019 | Manuscript Published on 22 June 2019 | PP: 473-478 | Volume-8 Issue-3S, February 2019 | Retrieval Number: C10990283S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Recent development in internet connectivity based mobile technologies leads a tremendous growth on Mobile Ad hoc Networks (MANETs). Further, these networks are specialized and serving the need of high-speed internet connection to the general user and for industries. Furthermore, the raise in the technology rises along with the security issues and threats. It is noticed that the attack over these MANET are high and severity of the attack launched over the networks are increasing drastically. An existing security system, which completely relies on the conventional signature based approach fails in detecting these attacks. Hence, an efficient classification model is required. Further, to create a robust and efficient model, this paper proposes an ensemble model, which holds the group of classifier for classification. The ensemble model validates the input in each classifier and finally ranks the accuracy. The best accuracy among the classifiers is taking into consideration. The entire experimental setup is experimentally tested and validated using a secure test bed with four machines running in Kali Linux operating system. From the experimental results, it is confirmed that the proposed ensembling model has an accuracy of more than 90% even for unknown attacks.
Keywords: Ensemble, Machine Learning, Wireless Attacks, MANET, Wireless Networks.
Scope of the Article: Mobile Adhoc Network