Network statistics in function of statistical intrusion detection

Abstract

Intrusion detection is used to monitor and capture intrusions (attacks) into computer network systems which attempt to compromise their security. A lot of different approaches exist for statistical intrusion detection. One of them is behavioral analysis, thus in accordance with this, a model-based algorithm is presented. The paper also elaborates the way of determining the control limits of regular traffic using descriptive statistics. In addition, the paper deals with the statistical formulation of relation between the average and maximum value of network traffic and discuss the time factor in intrusion detection. © 2010 Springer-Verlag Berlin Heidelberg.