An efficient attack detection for Intrusion Detection System (IDS) in internet of medical things smart environment with deep learning algorithm
Recently, the Internet of Things (IoT) has been an invention for the creation of intelligent worlds. IoT is considered a widely recognized implementation that includes intelligent health care, intelligent transport, and intelligent grids. In any technology depending on the IoT model, in which the In...
Saved in:
Main Authors: | , |
---|---|
Format: | Article |
Published: |
Little Lion Scientific
2023
|
Online Access: | http://psasir.upm.edu.my/id/eprint/106549/ https://www.jatit.org/volumes/hundredone11.php |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Recently, the Internet of Things (IoT) has been an invention for the creation of intelligent worlds. IoT is considered a widely recognized implementation that includes intelligent health care, intelligent transport, and intelligent grids. In any technology depending on the IoT model, in which the Internet of Medical Things (IoMT) is an important technique, privacy and secrecy are considered the major problems driven by numerous attacks triggered by intruders. The detection of unknown attacks is one of the main challenges in intrusion detection system (IDS). Researchers have performed multiple typing and detected anomaly traffic methods in the past decades without earlier understanding the attack signature specifically to the IoT environment. Therefore, an intrusion detection method for attacking and detecting anomalies in an IoT system must be enhanced. To achieve this, we measured the performance of three deep learning algorithms for normal and abnormal detection of IDS, and a comparison was made to select the best performance of the deep learning algorithm for detection in IDS, such as RNN, DBN and CNN. The CICIDS2017 dataset was used to analyze the performance of the existing intrusion detection system model. Additionally, the results of the deep learning algorithms will be evaluated using five confusion matrices, namely, accuracy, precision, recall, F1Score, and false-positive rate). It should be noted that the results showed a good average because most of them exceeded 90 of the total confusion matrix for all three deep learning algorithms that have been evaluated. |
---|