Comprehensive Analysis of Intrusion Prevention and Detection System and Dataset used in WSN using Machine Learning & Deep Learning

Abstract

The Wireless Sensor Networks (WSN), which are spatially dispersed small-sized low-power sensor devices with wireless radio transceivers that sense numerous physical events and gather data in a variety of situations. Because of their restricted capabilities, haphazard deployment, and unsupervised operations, in militant circumstances, like enemy zones, the sensor nodes were subject to a range of assaults can have their security broken. When sensor nodes are physically seized and changed during deployment in a hostile environment, WSNs are particularly vulnerable to DoS attacks. Almost every layer in WSNs is vulnerable to DoS assaults, which employ a number of attacks. In this paper, the characteristics of an effective intrusion prevention and detection system in wireless sensor networks are defined, and a detailed study of datasets used in Machine Learning (ML) and Deep Learning (DL) networks is made to see if Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS) can perceive and thwart DoS attacks with higher classification accuracy rates.