Investigation of RBFN application for anomaly-based intrusion detection on IoT networks

Abstract

The wide selection of current Internet of Things (IoT) worldview has prompted the innovation of savvy urban areas. That incorporates an unprecedented number of objects of every kind from smart microwaves to self-driving vehicles to wearable wellness devices. Sensitive information produced by these devices represents a critical test for manufacturers who are looking to completely shield their devices from various cyber-attacks. The IoT systems are developing exponentially and presenting new cybersecurity demonstrations since these IoT devices are related to sensors and these sensors are straightforwardly connected with large data servers. This research introduces an intelligent system to detect anomalies in IoT datasets to guard the security penetrations, created with Neural Network Technique, i.e., Radial Basis Function Network. This canny technique examines the anomalies and attacks in the IoT enabled systems. The proposed method attained a 99.3% test accuracy with 0.2% of the false-positive rate.