Self-similarity network traffic is considered as one of stochastic process studies in telecommunications engineering. In determining self-similarity traffic, Hurst value is an important parameter to be measured. This paper presents self-similarity traffic measurement using Rescaled Range, R/S statistical method in estimating Hurst parameter value. Inbound internet traffics on an IP-based campus network in Malaysia, which implements a 16.0 Mbps speed to internet and supports 10GE bandwidth at switch level, are captured and measured. The objectives of this research are to observe and present the existence level of Hurst parameter value, type of self-similarity and overall percentage of Hurts parameter estimation. The inbound traffic is measured due to its relevancy to next development on policing and shaping algorithm traffic model. Solarwinds Net Flow machine is setup on a campus gateway to its Wide Area Network (WAN). Data of the traffic like in flow, size and speed were taken over 20 days and 14 weeks in different inter-arrival time. These traffics are analysed, which lead to the impacts of packet loss, throughput and speed in network performance. Results present the Hurst parameter value, the existence of Long Range Dependant Self-similarity traffic distribution and percentage level of Hurst parameter value for the three types of captured traffic.