| Autor: |
Hanxi Zheng, Huanpu Yin, Haisheng Li |
| Jazyk: |
angličtina |
| Rok vydání: |
2024 |
| Předmět: |
|
| Zdroj: |
IEEE Access, Vol 12, Pp 149099-149114 (2024) |
| Druh dokumentu: |
article |
| ISSN: |
2169-3536 |
| DOI: |
10.1109/ACCESS.2024.3476136 |
| Popis: |
In recent years, the rapid growth of Internet of Things (IoT) devices has introduced significant security risks, emphasizing the need for effective IoT device management. Traditional device identification methods, such as traffic analysis and machine learning, which rely heavily on static features and prior knowledge, often struggle to maintain accuracy in dynamic IoT environments characterized by frequent changes and large-scale deployments. To address this challenge, we propose the Feature Extraction-Deep IoT (FE-DIoT) method to improve the accuracy and reliability of IoT device identification in large-scale networks by integrating a feature selection algorithm with a classification model. Our approach leverages the Dynamic Weight Adjustment-Based Recursive Feature Elimination (DWA-RFE) algorithm to effectively minimize redundancy, thus enhancing the robustness of the model. Additionally, we implement a Deep Cross Network with Feature Extraction (DCN-FE) module to improve device classification precision by extracting the most significant features from network traffic data. The experimental results in our dataset and other public datasets demonstrate that FE-DIoT significantly outperforms existing methods in terms of robustness and classification accuracy. These findings indicate that FE-DIoT provides a practical and effective solution to improve IoT device management and security in complex large-scale networks. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
