A Novel Hybrid Deep Learning Framework for Cyber-Attack Detection in IoT Networks

A Novel Hybrid Deep Learning Framework for Cyber-Attack Detection in IoT Networks

Velpula Sunilkumar, B.Jayasri, Bollaveni Renuka, Vorem Ravindar

Assistant Professor, Department of Computer Science & Engineering, Jyothishmathi Institute of Technology and Science, Telangana.

Abstract: The proliferation of Internet of Things (IoT) devices has introduced new vulnerabilities and expanded the attack surface of modern networks. Traditional security approaches often fall short due to the dynamic and heterogeneous nature of IoT environments. This paper proposes a novel hybrid deep learning framework that integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks for real-time cyber-attack detection in IoT networks. The proposed architecture leverages the spatial feature extraction power of CNNs and the temporal sequence modeling capabilities of LSTMs to accurately identify and classify various types of network intrusions. Evaluated on benchmark datasets such as UNSW-NB15 and BoT-IoT, the framework demonstrates superior accuracy, precision, recall, and F1-score compared to traditional machine learning and single deep learning models.

Keywords: Internet-enabled; reconfigurable wireless devices; spotted hyena optimizer; long short term memory; node MCU; multi-tier architecture