To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
CNN and RNN-LSTM Algorithms for the Detection Threats in IoT |
|
Author |
Nouf Bader Aljabri and Omar H Alhazmi |
| Citation |
Vol. 26 No. 4 pp. 111-120 |
|
Abstract |
Recently, the Internet of Things (IoT) had significant influence and much interest in many industrial sectors, including smart cities, healthcare fields, and automobiles. IoT devices generate a huge amount of data that gets transmitted across public networks. However, these devices are susceptible to various cyber threats due to their physical diversity and lack of security mechanisms. This research presents an IoT cyber threats detection system based on the Deep Learning (DL) model. DL has proven to be a powerful approach for effectively addressing and preventing cyberattacks on IoT infrastructure. The proposed system serves a critical role in protecting sensitive information by identifying and mitigating malicious activities. This paper focuses on the proposed detection system carrying out RNNs with LSTM and CNNs using the UNSW-NB15 dataset and the results show that the proposed system based on CNN and RNN-LSTM algorithms when including HashFeature significantly outperformed when including Principal Component Analysis (PCA) in all evaluation metrics. |
|
Keywords |
Internet of Things, Cyber Threats, CNN, RNN, LSTM, Attack detectio |
|
URL |
