An approach to measure the Hurst parameter for the Dhaka University network traffic

The main goal of this work was to analyze the network traffic of the University of Dhaka and find out the Hurst parameter to assess the degree of self similarity. For this verification a number of tests and analyses were performed on the data collected from the University Gateway router. The conclusions were supported by a rigorous statistical analysis of 7.5 millions of data packets of high quality Ethernet traffic measurements collected between Aug '07 and March'08 and the data were analyzed using both visual and statistical experimentation. Busy hour traffic and non-busy hour traffic, both were considered. All the software was coded using MATLAB and can be used as a tool to determine the inherent self similarity of a data traffic.