Network Intrusion Detection using Wide Tuned Convolutional Neural Network on IoT Network
Lia Hafiza, Reni Dyah Wahyuningrum, Syifa Maliah Rachmawati, Sherfina Salshabila · 2023
The Internet of Things (IoT) traffic is growing as a result of the enormous number of IoT devices. Intrusion detection systems (IDS) are necessary for detecting network threats. As a solution to the challenges, this research proposes a deep learning-based intrusion detection approach. A Wide Tuned Convolutional Neural Network (WTCNN) is designed to identify an attack in the real-time traffic of an IoT network environment. Data preprocessing and hyperparameter tuning are involved to achieve better performance. Based on result, the proposed method could achieve 0.99 of accuracy, 0.019 of loss, and 0.99 recall, precision, and F1 score. This proposed method gets the best result compared to existing algorithms.