Deep learning and layered architecture-based anomaly detection method for IoT

With the rapid development of the internet of things (IoT), the proliferation of IoT devices has led to massive data generation and transmission. However, security issues, especially anomaly detection, have become a major challenge. Traditional anomaly detection methods often rely on rule-based techniques or conventional machine learning models, which face issues such as low accuracy and high computational costs when handling large-scale, high-dimensional IoT data. To address these challenges, this paper presents a novel IoT anomaly detection method based on deep learning and hierarchical architecture (DeepIoT-HAD). This approach combines deep autoencoders (AE) with a hierarchical architecture to efficiently process the diversity and complexity of IoT data, improving detection accuracy while reducing computational resource consumption. Experimental results show that DeepIoT-HAD outperforms traditional methods and existing deep learning models in terms of detection accuracy and computational efficiency across multiple benchmark datasets.