Traffic modeling methodology for performance analysis of a class of wide-area network problems

Positive auto-correlations in input traffic dramatically increase the queue length statistics at a multiplexer. Further, traffic over several data networks has been found to possess a slowly decaying auto-correlation structure. These characteristics are difficult to capture using simple traditional models, and yet accurate studies of network environments need to be based on realistic traffic. This paper presents a traffic modeling methodology for network performance analysis that appears to preserve the temporal dependencies and the marginal distribution of the packet arrival process for connection-oriented traffic over a wide-area backbone network.