This study investigates the use of Convolutional Neural Networks (CNN) and Variational Autoencoders (VAE) for anomaly detection in IoT device traffic. The research aims to enhance the detection capability of security threats in IoT environments. The CNN model achieves an accuracy rate of 95.85% on the test dataset, effectively distinguishing between different types of IoT device traffic. The VAE model excels in anomaly detection by capturing abnormal patterns using reconstruction loss and KL divergence. The combined use of CNN and VAE models provides a comprehensive solution to cybersecurity challenges in IoT environments. Future research directions include exploring diverse IoT traffic data, practical deployment for validation, and further optimization of model structures and parameters to improve performance and applicability. The study also discusses the limitations of the current model, such as data imbalance and the need for better generalization to unseen IoT device traffic patterns.This study investigates the use of Convolutional Neural Networks (CNN) and Variational Autoencoders (VAE) for anomaly detection in IoT device traffic. The research aims to enhance the detection capability of security threats in IoT environments. The CNN model achieves an accuracy rate of 95.85% on the test dataset, effectively distinguishing between different types of IoT device traffic. The VAE model excels in anomaly detection by capturing abnormal patterns using reconstruction loss and KL divergence. The combined use of CNN and VAE models provides a comprehensive solution to cybersecurity challenges in IoT environments. Future research directions include exploring diverse IoT traffic data, practical deployment for validation, and further optimization of model structures and parameters to improve performance and applicability. The study also discusses the limitations of the current model, such as data imbalance and the need for better generalization to unseen IoT device traffic patterns.