| Title |
Privacy-Preserving Active Learning On The Internet Of 5G Connected Artificial Intelligence Of Things |
| ID_Doc |
43159 |
| Authors |
Ahmed U.; Lin J.C.-W.; Srivastava G. |
| Year |
2022 |
| Published |
IEEE Internet of Things Magazine, 5, 1 |
| DOI |
http://dx.doi.org/10.1109/IOTM.001.2100205 |
| Abstract |
With the advancement in 5G/6G networks, secure communication can help to develop data security applications. A layered network in the Artificial Intelligence of Things helps to assist data monitoring for the operation of critical components of smart cities. Unmanned aerial vehicles (UAVs) are often connected to the mobile network and are vulnerable to identify thefts and attacks. Safe UAV communication can help a number of data-driven applications to perform critical tasks. We introduce a deep reinforcement learning technique for sensitizing private information for a specific vehicle link across vehicular ad hoc networks. We use entropy-based active learning to classify the sensitive sensors data. The entropy-based active-learning-based method significantly increases training instances for the deep attention-based model. The proposed method helps introduce the new features based on the 5G/6G network while balancing safety, privacy, security, and confidentiality through a sanitization procedure. © 2018 IEEE. |
| Author Keywords |
|
