Abstract:
IoT technology and IoT smart devices are being used widely in recent years. As the devices and services are being developed with more sophisticated technologies, their se...Show MoreMetadata
Abstract:
IoT technology and IoT smart devices are being used widely in recent years. As the devices and services are being developed with more sophisticated technologies, their security is a prime factor to rely upon, in order to keep the services up and running properly. Among IoT network cyber attacks, DDoS attacks are prevalent. In order to detect and mitigate these cyber attacks, many machine learning based solutions are being implemented considering the kaggle datasets. this paper focus to detect DDoS attacks in IoT networks by using the proposed hybrid model of Jellyfish Swarm Optimization (JSO) and random forest (RF) algorithms using NS3. The DDoS scenario is designed and simulated in NS3 to generate the IoT DDoS dataset. JSO is bio-inspired algorithm used for feature selection while RF is machine learning used to detect DDoS. The proposed hybrid model improves the accuracy by 2 % compared with RF. Precision, Recall, f1-score statistics are also improved.
Published in: 2023 International Conference on Integrated Intelligence and Communication Systems (ICIICS)
Date of Conference: 24-25 November 2023
Date Added to IEEE Xplore: 08 February 2024
ISBN Information: