Enhanced Security in IoT Networks through Optimized Deep Learning-Based Intrusion Detection
Yash Anand, Saravanan Durai · 2025
The rapid growth of network services, driven by development of several network infrastructures led to improved complexity and challenges. To address increasing complexity and security challenges in Internet of Things (IoT) networks, this research proposes a novel Deep Learning (DL) based Intrusion Detection System (IDS) named DCNN-HGW-IMFO, which integrates a Deep Convolutional Neural Network (DCNN) with Hybrid Grey Wolf-Improved Moth Flame Optimization (HGW-IMFO) algorithm. The primary objective of this work to increase detection accuracy and reduce error rates in IoT environments. The proposed system leverages the DCNN’s ability to autonomously learn and extract high-level features from IoT traffic data, while the HGW-IMFO algorithm optimizes model parameters for improved detection performance. Data preprocessing steps such as cleaning, one-hot encoding, and normalization ensure quality input to the model. Experimental results using the UNSW-NB15 dataset show that the DCNN-HGW-IMFO system achieves an accuracy of 94%, precision of 94.1%, recall of 97.3%, and an F1-score of 95.7%. These outcomes confirm the effectiveness and robustness of the proposed approach in improving IoT security.