A detection of IoT based IDS attacks using deep neural network

The aim of the Internet-of-Things (IoT) is to seamlessly interconnect thousands or millions of physical sensors / devices by monitoring, analyzing and evaluating massive amounts of data gathered from interconnected IoT devices. The rapid increase in the usage of internet connectivity demands for hig...

Full description

Saved in:
Bibliographic Details
Main Authors: Sarika, Sabavath, Velliangiri, S., Ravi, M.
Format: Conference Proceeding
Language:English
Subjects:
Artificial neural networks
Datasets
Internet of Things
Malware
Telemetry
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The aim of the Internet-of-Things (IoT) is to seamlessly interconnect thousands or millions of physical sensors / devices by monitoring, analyzing and evaluating massive amounts of data gathered from interconnected IoT devices. The rapid increase in the usage of internet connectivity demands for high network security and especially IoT devices are easily compromised than computer systems that leads to an increase in the botnet attacks. In this paper, we propose deep autoencoders to detect suspicious network activity through vulnerable IoT devices. To Evaluate the proposed method, there will be no standard datasets are available for examine. However, we used MQTT (Message Queuing Telemetry Transport (MQTT) protocol based attacks. The dataset has provided with three levels of features such as packet-based, uni-directional and bi-dierction flow. The experimental results proved that performance of the proposed system has achieved high yield of accuracy and low false positive rate.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0057952