NETWORK TRAFFIC ANALYSIS USING QUEUING MODEL AND REGRESSION TECHNIQUE: A CASE STUDY OF FUTA NETWORK

Abstract

The increasing flow of network traffics on business and academic networks have raised the need for proper management of network traffic flow. Network Traffic analysis provides a way of determining network performance measures as it offers an insight to what type of traffic is flowing through a network. This research explores the analysis of internet traffic network of Federal University of Technology, Akure (FUTA) using M/M/1 queuing model and regression techniques. Traffic data flows were captured over a period of one month using Wireshark capturing tool at different strategic locations within the University. Parameter such as arrival rate and service rate is used to obtain the intensity of traffic at these locations. Traffic intensity greater than or approaching one Erlang means that the rate of packet arrival exceeds service rate, this indicates high traffic while value lesser than one Erlang indicates less traffic. From the result obtained, University locations such as Obakekere (school café) has the highest traffic intensity (0.95), followed by Obanla (department of computer science) with (0.87) while New Postgraduate Hostel has the least traffic intensity (0.20). Multiple Regression technique is used to determine the variability of the captured data, and the result obtained shows a good level of correlation between the independent variables and dependent variable. The captured data were analyzed using Statistical Package for Social Sciences (SPSS). This analysis provides methods for predicting the performance of the network, in order to make adequate provisions for future requirements needed to improve the service provided for users.