Open AccessAnomaly based Intrusion Detection using Neural Networks in 5G Network
Author(s) -
Thota Guna Durga Prashanth
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.38691
Subject(s) - computer science , intrusion detection system , identification (biology) , convolutional neural network , virtualization , network security , artificial neural network , computer security , malware , anomaly detection , software , artificial intelligence , data mining , cloud computing , operating system , botany , biology
Abstract: Ensuring the organizations of tomorrow is set to be a difficult space due to expanding digital protection dangers and enlarging assault surfaces made by the Internet of Things (IoT), expanded organization heterogeneity, expanded utilization of virtualisation innovations and circulated structures. This paper proposes SDS (Software Defined Security) which is a method gives mechanized, adaptable and versatile framework. SDS will tackle momentum progresses in AI to plan a CNN (Convolutional Neural Network) utilizing NAS (Neural Architecture Search) to distinguish irregular organization traffic. SDS can be applied to an interruption location framework to make a more proactive and start to finish protection for a 5G organization. To test this presumption, ordinary and irregular organization streams from a mimicked climate have been gathered and examined with a CNN. The outcomes from this strategy are promising as the model has recognized harmless traffic with a 100% exactness rate and irregular traffic with a 96.4% identification rate. This exhibits the viability of organization stream investigation for an assortment of normal pernicious assaults and furthermore gives a suitable alternative to discovery of encoded vindictive organization traffic. Keywords: 5G Security, IoT Security, Automated Intrusion Detection Systems, Convolutional Neural Networks, Artificial Intelligence, Software Defined Security