Multifractals in ATM traffic multiplexing

This paper discusses some of the issues encountered in the study of the multiplexing process in ATM. It has been observed that a self-similar process (second-order moment) describes ATM traffic fairly well. This also applies to the study of multiplexing but only in the stationary case. In the field, the traffic process is clearly non-stationary, and therefore there is a need for new analysis methodologies. We present that by observing higher-order moments it is possible to obtain a more general characterization of traffic and this can be used in the study of multiplexing.