Detecting Anomalous Network Traffic in IoT Networks

摘要: Network operators need effective tools to quickly detect anomalies in traffic data for identifying network attacks. In contrast traditional Internet, detection of anomalous IoT (Internet Things) networks is becoming a challenge task due limited resources and performance. Comprehensive methods are no longer networks, calling developing lightweight solutions. Principal Component Analysis (PCA) techniques can help reduce computing complexity, thus, anomaly based on PCA received lot attention the past. However, could not be directly applied with constrained This paper investigates detecting networks. We propose novel scheme two levels using techniques. The first level quick few principal components while second detailed number components. investigate selection parameters distance calculation formula several experiments show feasibility our proposed scheme.