Adaptive sampling technique for computer network traffic parameters using a combination of fuzzy system and regression model

In order to evaluate the effectiveness of wired and wireless networks for multimedia communication, suitable mechanisms to analyse their traffic are needed. Sampling is one such mechanism that allows a subset of packets that accurately represents the overall traffic to be formed thus reducing the processing resources and time. In adaptive sampling, unlike fixed rate sampling, the sample rate changes in accordance with transmission rate or traffic behaviour and thus can be more optimal. In this study an adaptive sampling technique that combines regression modelling and a fuzzy inference system has been developed. It adjusts the sampling according to the variations in the traffic characteristics. The method's operation was assessed using a computer network simulated in the NS-2 package. The adaptive sampling evaluated against a number of non-adaptive sampling gave an improved performance.