Lecture 12

# It is important to note that the poisson arrival can

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ly distributed with parameter . n = tn+1 , tn Prf n  sg = 1 , exp,s = 1 , e,s: 2) Sum of two independent Poisson processes is a Poisson process with rate being the sum of the two. This holds for more than two also. It is important to note that the Poisson arrival can also be described by a pure-birth homogeneous process. CS 522, v 0.94, d.medhi, W’99 14 9. Revisiting a network link In the section on M/M/1, we mentioned why the results are applicable for a data network link. We need to reiterate that M/M/1 results are applicable for Poisson arrival (discussed above) and exponentially distributed service time. Packet arrivals characterized by Poisson arrival is a reasonable assumption, although this is NOT always a good one. However, this helps us get started on getting a handle on doing some analysis. Typically, when we think of a network link we often have some idea about the rate such as 56 Kbps etc. This is a deterministic rate. Now, how do we get exponentially distributed service time? In...
View Full Document

## This document was uploaded on 03/19/2014 for the course CS 6030 at Western Michigan.

Ask a homework question - tutors are online