Detection of Network Attacks on Application Servers Using Deep Learning in IoT Environments
Niranjan W. Meegammana, Harinda Fernando · 2023
Internet of Things (IoT) comprises interconnected smart devices that collect data, control systems, are increasingly used in critical infrastructure, and raise security concerns due to their inherent vulnerabilities, where conventional security measures struggle to defend IoT networks. This research explores the use of Deep Learning for detecting network attacks on IoT application servers, addressing vital security concerns. The study, employing the Artificial Neural Network (ANN) DL model, achieved an impressive 93% accuracy and precision Score of 0.99, highlighting its robustness in identifying various network attack types on IoT application servers. Furthermore, it demonstrated strong performance in terms of precision, recall, and F1 scores, around 0.95, 0.92, and 0.93, respectively, showcasing the ANN model's ability to make precise predictions while minimizing false positives. These findings indicate that DL, especially the ANN model, significantly strengthens IoT security by safeguarding application servers from diverse security threats. Future work will focus on countering emerging zero-day attacks and further integrating the ANN model with application firewalls to enhance IoT security.