Detection and prevention in WSN security framework using deep learning against black hole and wormhole attacks
This paper proposed an energy-aware and adaptive IDS for WSNs with the main focus of tackling the Black Hole and Wormhole attacks on the routing systems. Based on the achieved results, the advanced feature generation is delivered by the use of the GANs, while the WOA is used for the tuning of the pa...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-10-01
|
| Series: | Ain Shams Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S209044792500365X |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper proposed an energy-aware and adaptive IDS for WSNs with the main focus of tackling the Black Hole and Wormhole attacks on the routing systems. Based on the achieved results, the advanced feature generation is delivered by the use of the GANs, while the WOA is used for the tuning of the parameters of the method; furthermore, deep Q-learning is used for the adaptive learning of the attacks’ patterns. Several simulation experiments were performed with the help of a MATLAB-based WSN environment. The presented hybrid model provided a 96.6 % detection accuracy, a recall of 0.803 and a specificity of 0.97, which were higher than those obtained from the traditional ML and the DL models. Based on the results achieved in this study, it is evident that the proposed IFM is highly scalable, extremely robust and suitable for real-time application in constrained WSNs. The memory usage of the system is approximately 930 MB, which renders the structure perfectly suitable for the implementation in the resource-constrained WSN settings. Future studies will involve the deployment of the testbed physically and the incorporation of energy-sensitive routing functions. |
|---|---|
| ISSN: | 2090-4479 |