Abstract

The proliferation of Internet of Things (IoT) devices has significantly expanded the attack surface, making IoT networks increasingly vulnerable to diverse cyber threats. Traditional intrusion detection systems (IDSs) often struggle to effectively identify sophisticated attacks in the dynamic and heterogeneous IoT environment. This paper proposes an adaptive intrusion detection system based on a hybrid deep learning model combined with feature engineering techniques to enhance cybersecurity in IoT networks. The proposed system leverages feature engineering to extract relevant and informative features from network traffic data, which are then fed into a hybrid deep learning model consisting of a Convolutional Neural Network (CNN) for feature extraction and a Long Short-Term Memory (LSTM) network for temporal pattern analysis. The CNN component automatically learns hierarchical features from the preprocessed data, while the LSTM network captures long-range dependencies in the sequential network traffic, enabling the system to detect complex and evolving attack patterns. The performance of the proposed system is evaluated using a publicly available IoT network traffic dataset. The experimental results demonstrate that the proposed system achieves superior detection accuracy, precision, recall, and F1-score compared to existing IDSs, highlighting its effectiveness in mitigating cyber threats in IoT environments. Furthermore, the adaptive nature of the system allows it to dynamically adjust its parameters and feature selection based on the evolving threat landscape, ensuring robust and reliable cybersecurity protection for IoT networks.