ANALYSIS OF PACKET LOSS PROCESSES IN HIGH-SPEED NETWORKS

The packet loss process in a single server queueing system with a finite buffer capacity is analyzed. The model used addresses the packet loss probabilities for packets within a block of consecutive sequence of packets. In contrast to other work that used an independence assumption to compute the loss probabilities of packets within a block, an analytical approach is presented that yields efficient recursions for the computation of the distribution of the number of lost packets within a block of packets of fixed or variable size for several arrival models and several number of sessions. Numerical examples are provided to compare the distribution obtained from our analysis with the distribution obtained by using the independence assumption. The results give insight to the following areas related to high-speed networks: 1) forward error correction schemes become less efficient due to the bursty nature of the packet loss processes; 2) real time traffic such as voice and video might be more sensitive to network congestion than was previously assumed, 3) the retransmission probability of ATM messages has been over-estimated by the use of independence assumption.