Multi-fractal analysis of IP-network traffic for assessing time variations in scaling properties

This paper presents a multi-fractal-based analysis of IP-network traffic in terms of the time variations in scaling properties. To obtain a comprehensive view in analyzing the scaling properties of IP-network traffic, we used a self-organizing map, which is an effective tool to map high-dimensional data onto a low-dimensional domain. Based on sequential measurements of IP-network traffic at two locations, we checked time variations in multi-fractal-based properties of measured data sets. In performing the self-organizing map-based analysis, we used three parameters: the highest value and the range of generalized fractal dimensions and the network throughput of measured network traffic. We visually confirmed that measured data sets could be classified and mapped in accordance with the network traffic properties, resulting in the combined depiction of the multi-fractal-related properties and network throughput, which can give us an effective assessment of network conditions at different times.