Get started Bring yourself up to speed with our introductory content.

G723.1 basics

From where and on what basis does codec G723.1 get commands to use optional features such as post filtering, bit rate change, high pass filtering, etc. in VoIP applications.

In answer to your question, G732.1 is a standard that falls under the ITU-T, headquartered in Geneva, Switzerland. This is an international organization much like IEEE, TIA, and others, but they specifically address global telecommunication networks and services. They are the keeper of the standards and addendums for all of the compression schemes to assure that G732.1 in one country can be understood by the same in another country.

I sit on several of the standards boards, and the entire process would impress you. Once a standard is released, there are often several vendors with solutions that already comply due to their work on the standards boards and knowing the direction in which the standard will go. It is VERY thorough and participation is encouraged from around the world.

You can look at a lot of the information on these standards at www.itu.int

Further, implementation of these features is based on several things... congestion, throughput, and equipment connected to your system. These things vary with different equipment manufacturers as they all use their own buffering and forwarding techniques. Is there a particular problem you are trying to solve with a particular vendor's equipment. That would give me a bit more information to answer this question, but I will answer what I think you are asking. Feel free to ask again, if I don't hit the mark. Video and Voice traffic needs to move at a higher priority than regular network traffic. So, for instance, if you look at a typical voice call, the voice conversation is sampled (pieces of the conversation are taken) and placed in packets. Video is the same. The higher the sampling rate, the greater the quality of signal, but the lower the compression. Bandwidth is far greater for a fully sampled signal. That said, the human brain can fill in the blanks for video and audio signals so that they remain audible.

If you have ever looked at a digital TV signal when the picture gets snowy or turns to blocks, that is an example of lost packets on the network. In Video transmission, the first frame (index frame) is fully sampled and subsequent frames only transmit differences from the index frame (lossy compression).

In voice, the samples are re-assembled at the receiving end.

If you are having problems with quality, then you will want to do one of two things. First, look at your network. A good analyzer will help, and find out if you are having lots of errors or discards. If so, fix this first.

If not, then you can experiment with the sampling rate. G723.1 allows for either 5.3kbps or 6.4kbps. Alternatively, depending on your equipment, you may just be able to change the priority of your voice or video packets to put them ahead of regular network traffic. How you do that depends on whether you are using traffic shapers, Layer 3 capabilities, Diffserve, QoS, etc. I hope this answers your questions. There are some whitepapers on www.siemon.com that discuss these issues.

This was last published in August 2003

Dig Deeper on Network protocols and standards