Hybrid Convolutional Neural Network for Robust Attack Detection in Wireless Sensor Networks

ABSTRACT In this article, we propose an intelligent hybrid model that combines deep learning with optimization techniques to enhance the security of wireless sensor networks (WSN) by detecting and preventing cyberattacks. In the rapidly expanding Internet landscape, security has become crucial due to the surge in Internet users. Various intrusion detection systems have been developed over time to detect and identify intruders using data processing methods. However, existing systems often fall short in achieving high detection accuracy. To address this, the article introduces an Enhanced Black Widow Optimization (EBWO) algorithm integrated with a Bidirectional Gated Recurrent Unit (BiGRU) and Attention Mechanism (ATTN) model for detecting malicious activities in IoT‐based WSNs. The EBWO algorithm optimizes the parameters of the BiGRU‐ATTN network, enhancing its performance. The proposed model's effectiveness is evaluated using various metrics and benchmarked against other popular deep learning algorithms. An experimental study is conducted using the WSN‐DS dataset to demonstrate that the proposed approach detects cyber threats in WSNs better than any other approach.

Перевод пока недоступен