Artificial Intelligence Techniques for Dynamic Path-Controllable Deep Unfolding Network to Predict the K-Barriers for Intrusion Detection using Wireless Sensor Networks: Trends and Challenges

Abstract

Wireless Sensor Networks (WSNs) have become essential for surveillance, border monitoring, and other security-critical applications, but their distributed nature, limited resources, and exposure to hostile environments make them highly vulnerable to intrusions and cyberattacks. Intrusion Detection Systems (IDS) play a vital role in maintaining network reliability and ensuring real-time threat mitigation. In recent years, artificial intelligence techniques, particularly deep learning and optimization-based models, have significantly enhanced IDS performance in WSNs. A key concept in this domain is K-barrier prediction, which determines the number of disjoint sensing barriers required to detect intrusions effectively, thereby improving coverage reliability. Advanced models such as Artificial Neural Networks (ANNs), Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), and hybrid approaches have been widely used for accurate prediction. The integration of Dynamic Path-Controllable Deep Unfolding Networks further improves performance by combining optimization with neural learning for adaptive and efficient inference. This review highlights recent advancements, compares methodologies, and identifies challenges such as energy efficiency, scalability, and real-time deployment, while suggesting future directions like edge intelligence and explainable AI.