Telecommunications and Teletraffic Theory
Presenter: Dr Andrew Coyle, TRC
Dates: please enquire
Venue: please enquire

About Telecommunications & Teletraffic Theory

The three-day short course “Telecommunications & Teletraffic Theory” is designed to introduce participants to fundamental methods used in the modeling and analysis of communication networks. The first two days are structured to deliver the basic knowledge behind Teletraffic Engineering. These are split into morning and afternoon lecturing sessions, which are accompanied by exercises and experiments. The third day is structured to encourage discussion on the modern challenges and issues facing Teletraffic engineers. It consists of more interactive lecturing sessions and no exercises nor experiments.

The material will consist of a set of course notes, a separate set of pre-reading notes, the TRC Traffic Analysis Tool (written in Java), a collection of octave/Matlab programs, some Windows programs, and a further set of exercises that can be performed in Excel. 

Course Topics 

Day 1: Classical Teletraffic Models

The course starts with a review of the classical queueing models and theory that have proved to be of great use in the analysis of traditional telecommunication systems, and are still relevant for modern day practice.

Basic Queueing Theory

  • Birth and death processes
  • The M/M/N/M queue
  • The Erlang loss formula
  • The Erlang C formula
  • Equilibrium distributions, as seen by arrivals
  • Poisson Arrivals See Time Averages (PASTA)
  • Little's law
  • Waiting times
  • Computer experiments on Teletraffic Models using the TRC Traffic Analysis Tool
Networks
  • Circuit switched networks: Exact calculation of blocking probabilities; Approximate methods: the Erlang Fixed Point Method (EFPM).
  • Tutorial on the EFPM using the TRC Traffic Analysis Tool
IP Network Routing
  • Routing overview - OSPF
  • Flow Controls within TCP
  • Basic models of TCP throughput

Day 2: Models for Data Networks

Here we look at the more recent developments in teletraffic theory and their relevance for data networks.

Modelling data streams

  • Timescale issues (call, session, packet).
  • Markovian arrival streams: Interrupted Poisson Processes (IPP); Switched Poisson Processes (SPP); Markov Modulated Poisson Processes (MMPP)
  • Computer generation of Markovian arrival streams using software
  • Modelling traffic streams using a Markovian model
Internet Traffic Characterization
  • Dependent arrival streams: Long range dependence (LRD) and Self-similarity; Fractional Brownian motion; Fractional ARIMA process
  • Computer generation of LRD and Self-similar processes
Basic Traffic Quality of Service
  • Traffic Shaping and Marking – Leaky Buckets etc.
  • Priority Queueing
  • DiffServ Capabilities

Day 3: Modern Teletraffic Engineering

In this session we discuss some of the more contentious and interesting issues in Teletraffic Engineering. The sessions will be significantly less technical and more interactive in nature.

Traffic Engineering

  • Objectives
  • Traffic Path setup using MPLS
  • Traffic apportioning using Constraint Based Routing
  • Experiments in Traffic Engineering
Origin-Destination (OD) Traffic Modelling
  • Traffic streams in IP Networks
  • Measuring the Network
  • Gravity Models to create Origin Destination traffic approximations
  • Simple Gravity Model experiments ·        
  • Enhanced models
 

For further information please contact:
Anne-Marie Eliseo
Industry Education Manager
phone: +61-8-8302-3928
email: industryeducation@nicta.com.au

NICTA is able to deliver in-house training.