Teletraffic Engineering

Introduction
Traffic engineering uses statistical techniques such as queuing theory to predict and engineer the behaviour of telecommunications networks such as telephone networks or the Internet.
The field was created by the work of A. K. Erlang in whose honour the unit of telecommunications traffic intensity, the Erlang, is named. The derived unit of traffic volume also incorporates his name. His Erlang distributions are still in common use in telephone traffic engineering.
The crucial observation in traffic engineering is that in large systems the law of large numbers can be used to make the aggregate properties of a system over a long period of time much more predictable than the behaviour of individual parts of the system.
The queueing theory originally developed for circuit-switched networks is applicable to packet-switched networks.
The most notable difference between these sub-fields is that packet-switched data traffic is self-similar. This is a consequence of the calls being between computers, and not people.

Teletraffic Engineering

Teletraffic engineering is the application of traffic engineering theory to telecommunications. Teletraffic engineers use their basic knowledge of statistics including Queueing theory, the nature of traffic, their practical models, their measurements and simulations to make predictions and to plan telecommunication networks at minimum total cost. These tools and basic knowledge help provide reliable service at lower cost. Because the approach is so different to different networks, the networks are handled separately here: the PSTN, broadband networks, mobile networks, and networks where the possibility of traffic being heavy is more frequent than anticipated