A fast method for real-time anomaly detection using routing statistics
Graham Booker · 2008
The detection of network anomalies is an important tool for network managers and service providers. Network anomalies can indicate the onset of a new virus or a directed malicious attack. In this paper, we present a fast method for detecting network anomalies at a router. Our algorithm can be easily integrated into a router's preexisting firmware with minimum modifications. In addition, its low memory and CPU requirements facilitate real-time anomaly detection.