Analysis and modelling of a broadband fiber access network with high peer-to-peer traffic load

This paper provides a statistical characterization of the traffic in a fiber network for voice and data transmission. This research operates with real data from more than 33,000 users. The most significant applications (ftp, http, p2p services, …) have been processed. Also, a model of the traffic generated is implemented, representing the usage that network subscribers make of the system. We demonstrate that this traffic is statistically self-similar, such behaviour having serious implications for the design, control and analysis of high-speed networks. The results in these processes have been validated using the real data provided by the fiber network operator Telecable Asturias, SAU.