A comparative study of the statistical methods suitable for network traffic estimation
Iarina Marian, Vasile Teodor Dădârlat, Bogdan Iancu · International Conference on Communications · 2009
Predicting network traffic has a great importance for many real time and non-real time applications, for planning the network resources and for traffic matrix computations. In this paper several estimation methods for IP network traffic are studied. Methods for both Short-Range Dependence (SRD) and Long-Range Dependence (LRD) characteristics are presented, and also offline and online prediction algorithms are described with their advantages and disadvantages and suitable practical applications.