Enhancing DoS Detection in WSNs Using Enhanced Ant Colony Optimization Algorithm

The increasing use of Wireless Sensor Networks (WSNs) is leading to network traffic growth as data exchange among WSN nodes increases. Protecting WSNs from Denial of Service (DoS) attacks is essential for enhancing data security and avoiding interruptions that can harm productivity and reputation. Detecting a DoS attack quickly is crucial to minimize its impact on the targeted system or network. To meet this requirement, it is critical to have an effective DoS attack detection mechanism that ensures system or network availability and safeguards data and resources. The suggested approach focuses on enhancing DoS attack detection, reducing anomalies, and offering an efficient way to protect WSNs from DoS attacks. A new framework has been proposed to improve DoS attack detection by using optimization techniques and Machine Learning (ML) algorithms to detect and manage DoS attacks effectively. This system integrates Ant Colony Optimization (ACO) with ML algorithms to propose the Enhanced Ant Colony Optimization (EACO) technique. The proposed system has been compared to ACO through the Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) algorithms to assess their performance in identifying DoS attacks. The results from the assessment show that when the EACO algorithm is paired with ML algorithms, it can achieve accuracy, sensitivity, specificity, and F1 scores between 87.6% and 99.8%. Furthermore, the EACO surpasses ACO in terms of accuracy, sensitivity, specificity, F1 score, precision, and Negative Predictive Value (NPV) by about 3.64%, 38.6%, 1.11%, 27.53%, 16.35%, and 2.78%, respectively.