Survey of AI and Optimization Methods for DoS Attack Detection in WSNs

The use of Wireless Sensor Networks (WSNs) in modern IoT applications is very important, as they can monitor and transmit data in different environments. Although their open communication nature, limited energy, and computational capacity render them highly vulnerable to Denial-of-Service (DoS) attacks that will deteriorate network performance and reliability significantly. This survey covers all aspects of recent developments using artificial intelligence (AI) based techniques for the detection and prevention of DoS attacks in wireless sensor networks (WSNs). It gives particular attention to the integration of deep learning models and metaheuristic optimization strategies in terms of making their detection accuracy better, minimizing predication of false positives, and improving the model’s adaptability in the way of constraints of resource limits. Real time DoS detection has been studied under the perspective of different ensemble learning strategies, hybrid deep learning frameworks, and feature selection methods. The paper also examines the efficiency of the dimensionality reduction and heuristic tuning techniques in improving model performance. It also discusses future research directions, challenges and open issues that are meant for guiding future innovation in this domain.