Theoretical mean-variance relationship of IP network traffic based on ON/OFF model

Abstract

Mean-variance relationship (MVR), nowadays agreed in power law form, is an important function.It is currently used by traffic matrix estimation as a basic statistical assumption. Because all the existing papers obtain MVR only through empirical ways, they cannot provide theoretical support to power law MVR or the definition of its power exponent. Furthermore, because of the lack of theoretical model, all traffic matrix estimation methods based on MVR have not been theoretically supported yet. By observing both our laboratory and campus network for more than one year, we find that such an empirical MVR is not sufficient to describe actual network traffic. In this paper, we derive a theoretical MVR from ON/OFF model. Then we prove that current empirical power law MVR is generally reasonable by the fact that it is an approximate form of theoretical MVR under specific precondition, which can theoretically support those traffic matrix estimation algorithms of using MVR. Through verifying our MVR by actual observation and public DECPKT traces, we verify that our theoretical MVR is valid and more capable of describing actual network traffic than power law MVR.traffic matrix, mean-variance relationship, self-similar

Network traffic simulation based on Network Processor

“…The validation of the model is also proved in an empirical way. For future, more statistical characteristics of network traffic will be introduced into the NTS-NP platform, for example the mean-variance relationship [6] .…”

“…However, so far as we learn the simulation approach of network traffic doesn"t catch up with the network technology development. We believe a feasible and reliable network traffic simulation method or tool should 1) implement the ordinary link or network properties such as latency, jitter, packet loss and fragmentation, 2) provide high bandwidth that matches the requirement of the target network or system and 3) have the same statistical character as the real network traffic, for example the self-similarity [1] and mean-variance relationship [6] . In this paper, we propose our work on Network Traffic Simulation based on Network Processor which successfully fulfills the requirements mentioned above.…”

Combinatorial Methodology Study for Network Traffic Prediction

“…Traditional network traffic prediction models include linear regression analysis, gray model and time series, etc. [2,3], these methods are based on linear modeling, but network traffic is a complex nonlinear system, therefore, traditional linear model cannot describe the characteristics of modern changes in network traffic accurately, prediction results are unreliable [4]. In recent years, with the development of nonlinear theory, nonlinear network traffic prediction models have appeared based on neural networks, supporting vector machines [5].…”

SPN-Based Performance Evaluation for ON-OFF Voice Traffic System

“…Nowadays, methodologies of performance analysis mainly take three ways, which are direct measurement [1], mathematic modeling [2], and simulation. Mathematic modeling and simulation are undoubted useful methodologies.…”