An aggregated statistical approach for network flood detection using Gamma-Normal mixture modeling

摘要: Abstract In this paper, we propose a fast statistical anomaly detector at the aggregated-level for two types of anomalies: floods and flash crowds. The performance detectors is significantly dependent on accuracy modeling. Thus, initially introduce new efficient model network traffic called Gamma Normal mixture (GNM). We study compatibility GNM using different tests. Consequently, design novel based generalized likelihood ratio test (GLRT) GNM. Moreover, to more accurately determine position anomalies, overlapped sliding windows have been applied in aggregation step. To evaluate proposed detector, use receiver operating characteristics (ROC). Experimental results under public traces, confirm high efficiency method. Also, comparison with its nearest competitor verifies higher lower computational load utilizing strategy.