A Neuro-Swarm Optimized Deep Convolutional Framework for Resilient and Energy-Aware DDoS Attack Detection in WSNs

Abstract

Distributed Denial-of-Service (DDoS) attacks have become a critical security challenge in intelligent Wireless Sensor Networks (WSNs) due to their ability to exhaust node energy, disrupt communication stability, and degrade network lifetime. Existing intrusion detection systems often suffer from high false alarm rates, unstable convergence behavior, and inefficient handling of high-dimensional traffic features. To address these limitations, this study proposes a Neuro-Swarm Optimized Deep Convolutional Framework (NSO-DCF) for Resilient and Energy-Aware DDoS Attack Detection. The primary objective of the proposed framework is to achieve accurate and low-latency DDoS detection while simultaneously improving energy efficiency and network lifetime under dynamic attack conditions. The proposed model integrates a one-dimensional Convolutional Neural Network (1D-CNN) for automatic deep feature extraction with Harris Hawks Optimization (HHO) for adaptive hyperparameter tuning, including filter size, learning rate, dropout ratio, and convolution depth. The framework was evaluated using the NSL-KDD dataset for benchmark intrusion analysis and the CIC-IDS dataset for real-time attack generalization assessment. Comparative analysis was conducted against TVCDNN-CUL and PSO-LSTM models to validate detection robustness and optimization efficiency. Experimental implementation and performance evaluation were carried out using Python, TensorFlow, Scikit-learn, and Google Colab simulation environments. The proposed NSO-DCNN framework achieved 95.8% accuracy, 95.4% precision, 95.1% recall, 95.2% F1-score, 94.8% energy efficiency, and 95.6% network lifetime improvement, while maintaining a low false alarm rate of 4.2%. The novelty of the study is to apply swarm-guided adaptive convolutional neural network optimization and energy-efficient intrusion detection system together for robust and energy-efficient DDoS classification in diverse cyber systems.