No abstract

Abstract-A testbed is described which allows both physical layer errors to be observed and analysed, as well as monitoring network performance via frame loss. Real network traffic loads can be used for testing, so that all measurements taken are representative of what would be seen in a deployed system. We illustrate our testbed with an examination of the behaviour of a well-known networking standard, Gigabit Ethernet, in conditions of reduced receiver power on optical fibre. Our testbed results show that the line codes used to represent the data in the network affect the bit error rate for that data. Along with the previously reported result that bit error rate and packet error rate have only a weakly deterministic relationship, this highlights the need for testing of all network layers within a complete system carrying real world traffic.

Abstract-A physical layer coding scheme is designed to make optimal use of the available physical link, providing functionality to higher components in the network stack. This paper presents results of an exploration of the errors observed when an optical Gigabit Ethernet link is subject to attenuation. The results show that some data symbols suffer from a far higher probability of error than others. This effect is caused by an interaction between the physical layer and the 8B/10B block coding scheme. We illustrate how the application of a scrambler, performing datawhitening, restores content-independent uniformity of packetloss. We also note the implications of our work for other (N,K) block-coded systems and discuss how this effect will manifest itself in a scrambler-based system. A conjecture is made that there is a need to build converged systems, with the combinations of physical, data-link, and network layers optimised to interact correctly. In the mean time, what will become increasingly necessary is both an identification of the potential for failure and the need to plan around it.