Advanced intrusion detection in IoT-enabled SDWSNs using artificial intelligence and feature optimization techniques
TL;DR
This paper proposes a hybrid IDS framework that integrates three key components, such as the novel exponential grey wolf-optimized grid search algorithm (EGWOGSA), which is a stochastic optimization algorithm, while feature selection is achieved using the novel symmetric gradient Boruta for enhanced feature selection algorithm (SGBFSA).
Abstract
The widespread deployment of the Internet of Things (IoT) and software-defined wireless sensor networks (SDWSNs) gave rise to new opportunities for smart environments but has also made these systems highly vulnerable to diverse cyberattacks. Conventional intrusion detection systems (IDSs) in IoT-enabled SDWSNs face major challenges such as high-dimensional data, inadequate preprocessing, feature extraction complexities, intricate feature selection, inefficient intrusion recognition, high computational cost, and real-time classification problems, which limit their performance in resource-constrained IoT networks. To address these limitations and challenges, this paper proposes a hybrid IDS framework that integrates three key components, such as the novel exponential grey wolf-optimized grid search algorithm (EGWOGSA), which is a stochastic optimization algorithm, while feature selection is achieved using the novel symmetric gradient Boruta for enhanced feature selection algorithm (SGBFSA), and a gated bidirectional recurrent convolutional neural network (GBR-CNN) algorithm for intrusion recognition. Extensive simulations, using the network simulator 3 (NS-3), were conducted with the NSL-KDD dataset to evaluate the framework for IoT-enabled SDWSNs. Results demonstrate that the proposed method outperforms other state-of-the-art models across most metrics, achieving 96.7% training accuracy and 91.6% testing accuracy, with 98.03% precision, 95.9% recall, and a 96.9% F1-score, and demonstrating low energy consumption and latency, high throughput, and a reliable packet delivery ratio (PDR), leading to an extended network lifetime. The study demonstrates that the proposed framework is suitable for real-time deployment. This research contributes to the advancement of security in IoT-enabled SDWSNs by proposing an efficient, accurate, and scalable IDS framework for securing them against evolving threats through advanced feature optimization and artificial intelligence techniques.
