New speech traffic background simulation models for realistic VoIP network planning

In order to overcome the known challenges of transmitting multimedia traffic over a switched packet network (i.e. latency, jitter, packet loss, etc.), careful network planning needs to take pace. Existing simulation platforms, particularly for Voice over IP (VoIP) simulations, have a limited selection of speech encoding algorithms. The primary objective of this paper is the creation of simulation models to be essential components of a simulation platform. Such a tool is aimed at supporting the planning and design phases of packet switched networks carrying voice traffic while considering realistic and current network conditions and simulation features. More specifically, the contribution of this paper is the creation of a speech background traffic generation model that generates traffic that follows statistical behaviour of a number of speech encoding algorithms. The purpose of such model is to provide relevant and current background traffic shape to VoIP simulations. To date, fix and variable data rate encoding algorithms (G.729, G.711, iLBC, Speex and AMR) are included in the codec choice for the model. Finally, the model and all its components are also made available to the scientific community.