Performance Evaluation of Machine Learning Techniques for DOS Detection in Wireless Sensor Network

The nature of Wireless Sensor Networks (WSN) and the widespread using WSN introduce many security threats attacks. An effective Intrusion Detection System (IDS) should be used to detect Detecting such an attack is challenging, especially detection Denial Service (DoS) Machine learning classification techniques have been as approach for DoS detection. This paper conducted experiment Waikato Environment Knowledge Analysis (WEKA)to evaluate efficiency five machine algorithms detecting flooding, grayhole, blackhole, scheduling at attacks in WSNs. evaluation based on a dataset, called WSN-DS. results showed that random forest classifier outperforms other classifiers with accuracy 99.72%.