Queue Analysis for Wireless Packet Data Traffic

Recent research based on Internet traffic measurements shows that the traffic flows have a fractal nature (i.e., self-similarity property), which causes the underestimation of the network engineering parameters when using the Poisson model. Preliminary field measurements demonstrate that packet data traffic in wireless communications also exhibits self-similarity. In this research, we investigate the queuing behavior of a self-similar traffic flow for wireless packet data applications. The traffic is modelled by an on-off source with heavy-tailed on periods. We derive a new relationship among the asymptotic distribution of the queue length, traffic specifications, wireless channel characteristics, and transmission rate. Computer simulation results are given to demonstrate the accuracy of the theoretical analysis.