An Efficient and Hybrid Deep Learning-Driven Model to Enhance Security and Performance of Healthcare Internet of Things
The development of wireless communication technology has led to an exponential growth in the Healthcare Internet of Things (H-IoT). Sensors and actuators are used in smart medical devices to collect data about the human body, which is then sent to the fog layer for analysis. However, H-IoT devices p...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10858121/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The development of wireless communication technology has led to an exponential growth in the Healthcare Internet of Things (H-IoT). Sensors and actuators are used in smart medical devices to collect data about the human body, which is then sent to the fog layer for analysis. However, H-IoT devices present security and Quality of Service (QoS) concerns because of their critical nature, complexity, and dynamic features, which make them incompatible with conventional network topologies. Furthermore, reducing superfluous data and identifying effective fog nodes are difficult tasks. We provide a novel Software-Defined Networking (SDN)- driven Deep Learning (DL) approach to develop a secure, intelligent, and efficient framework for smart H-CIoT networks in order to address the aforementioned challenges. In this method, we have first considered SDN architecture as a promising solution since it allows for reconfiguration over static network infrastructure and manages the distributed architecture of intelligent H-CIoT networks by keeping the data and control planes apart. Secondly, a security module based on Bidirectional Long Short-Term Memory (BiLSTM) is implemented to recognize various forms of attacks within the H-CIoT network. Third, the past medical records of the patients are used to train the DL model. It then makes an informed decision about whether to send the data to the fog layer. The CNN approach is also included in the suggested framework to choose the best fog node. The simulation results indicate that the proposed framework achieved an accuracy of 99.59%, an F1-score of 99.53%, a latency of 3 ms, energy consumption of 55 W, and a probability of 0.92%. It outperforms the baseline and current methods with improvements of 5% in accuracy, 4% in F1-score, 10 ms in latency, 25 W in energy consumption, and 0.66% in probability, respectively. |
|---|---|
| ISSN: | 2169-3536 |