A codec is a device capable of performing encoding and decoding on a digital signal. Each codec provides a different level of speech quality. The reason for this is that codecs use different types of compression techniques in order to require less bandwidth. The more the compression, the less bandwidth you will require. However, this will ultimately be at the cost of sound quality, as high-compression/low-bandwidth algorithms will not have the same voice quality as low-compression/high-bandwidth algorithms. The following table shows three standard codec types along with their corresponding Coder Type, Bit Rate, Frame Size Delay, and Look Ahead Delay. These particular three were chosen because these are the ones we will use in this lab, but there are others. ITU Recommended Delay Values/G.114
In this lab, you will build a wide-area network that will consist of several LANs spread across the country and connected via the Internet with two types of traffic: data and voice. You will configure three scenarios where the data and voice traffic generated will be held constant. The only parameter that you will alter will be the type of codec used. In the first scenario, the voice traffic will use G.711 (PCM); in the second scenario, the voice traffic will use G.729 (CS-ACELP); and the third scenario, the voice traffic will use G.723.1 (ACELP).
By holding the traffic generated for the data and voice users constant, and only changing the codec values, we will be able to see how the varying bit rate of the specific codecs used will affect the end-to-end delay for the data traffic, the end-to-end delay for the voice traffic, and the packet delay variation for the voice traffic for a wide-area environment. Motivation
Deciding on what codec to use is a tradeoff between bandwidth and speech quality; the more bandwidth the codec uses, the better the speech quality will be. Here are the codecs we will use, along with their corresponding MOS values. Codes