An intelligent hybrid deep learning approach for securing internet of things threat networks from cyber threats

Abstract

The rapid growth of internet of things (IoT) devices has increased their vulnerability to cyber-attacks, creating a need for accurate and adaptive intrusion detection systems (IDS). This study proposes a hybrid approach that integrates self-organizing maps (SOMs), deep belief networks (DBNs), and autoencoders to detect both known and unknown attacks in IoT networks. particle swarm optimization (PSO) is employed to optimize model parameters and improve detection accuracy and convergence. The proposed method was evaluated using synthetic and real-world network traffic datasets, including NSL-KDD, UNSW-NB15, and CICIoT2023. Standard cybersecurity performance metrics were used to assess the effectiveness of the model. Experimental results demonstrate an accuracy of 99.99% and Matthews correlation coefficient (MCC) values exceeding 99.50%. The results also indicate strong generalization across different attack scenarios. Overall, the proposed PSO-optimized hybrid model provides an accurate, adaptive, and robust solution for detecting diverse and emerging cyber-attacks in IoT networks.