<![CDATA[Wireless Sensor Networks (WSNs) have rapidly expanded across various applications. However, WSNs are highly vulnerable to attacks, one of which is Distributed Denial of Service (DDoS) attacks, posing a significant threat to many WSN systems. Therefore, a security technique is needed to protect WSNs from such attacks. This study aims to effectively detect DDoS attacks on WSNs using Machine Learning (ML), specifically with the Support Vector Machine (SVM) method. The testing in this study employs precision, recall, F1-score, confusion matrix, and ROC curve metrics. The results indicate the best performance in detecting Grayhole attacks. The Precision, Recall, and F1-score metrics for these attacks are all 1.0, meaning that Grayhole attacks are perfectly detected by the intrusion detection system utilizing the SVM method. Furthermore, based on the confusion matrix results, the system successfully identified Grayhole attacks with a Precision value of 99.7% and a Recall of 99.6%, indicating that only a few error detections occurred. However, the AUC test results for DoS attacks yielded a score of 0.5, suggesting that the system is unable to detect DoS attacks effectively.]]>
Copyrights © 2024
