sis 18: e49

Research Article

An Efficient Neuro Deep Learning Intrusion Detection System for Mobile Adhoc Networks

Download204 downloads
  • @ARTICLE{10.4108/eai.4-4-2022.173781,
        author={N. Venkateswaran and S. Prabaharan},
        title={An Efficient Neuro Deep Learning Intrusion Detection System for Mobile Adhoc Networks},
        journal={EAI Endorsed Transactions on Scalable Information Systems: Online First},
        volume={},
        number={},
        publisher={EAI},
        journal_a={SIS},
        year={2022},
        month={4},
        keywords={Deep Learning, Intrusion Detection, Mobile Adhoc Networks, MANET, Deep Neural Network, recurrent neural networks, intrusion detection systems, IDS},
        doi={10.4108/eai.4-4-2022.173781}
    }
    
  • N. Venkateswaran
    S. Prabaharan
    Year: 2022
    An Efficient Neuro Deep Learning Intrusion Detection System for Mobile Adhoc Networks
    SIS
    EAI
    DOI: 10.4108/eai.4-4-2022.173781
N. Venkateswaran1,*, S. Prabaharan2
  • 1: Associate Professor, Jyothishmathi Institute of Technology and Science, Karimnagar, Telangana, India - 505001.
  • 2: Professor in CSE, Mallareddy Institute of Engineering and Technology, Secunderabad, Telangana, India - 500055.
*Contact email: venkywn@gmail.com

Abstract

As of late mobile ad hoc networks (MANETs) have turned into a very popular explore the theme. By giving interchanges without a fixed infrastructure MANETs are an appealing innovation for some applications, for ex, reassigning tasks, strategic activities, nature observing, meetings, & so forth. This paper proposes the use of a neuro Deep learning wireless intrusion detection system that distinguishes the attacks in MANETs. Executing security is a hard task in MANET due to its immutable vulnerabilities. Deep learning gives extra security to such systems and the proposed framework comprises a hybrid conspiracy that joins the determination and abnormality-based methodologies. Executing the partial IDS utilizing neuro Deep learning improves the identification rate in MANETs. The proposed plan utilizes deep neural networks and a cross breed neural system. It demonstrates that Recurrent neural networks can successfully improve the identification and diminish the rate of false caution and failure.