Board Transcoding Market Needs and Solutions

Board Transcoding Market Needs and Solutions
Alan Percy
Director of Business Development, Boards Business Line

The Board Business Line Business Development team at AudioCodes has responded to an interesting opportunity in the carrier and service provider marketplace surrounding the need for transcoding resources and a “transcoding gap”. To help our customers and partners better leverage this opportunity, we’ve prepared what we hope is a useful list of frequently asked questions and answers:

What is transcoding?
In all Voice over IP solutions, the conversation is transferred in packets or small batches of digitized sound. There are a number of ways to represent human speech digitally, using coding algorithms. Common voice coding algorithms are illustrated below:

Transcoding is the conversion from one of the formats to another. In most cases, the conversion is done on-the-fly in real-time. Because many of the voice coding algorithms are fairly complex, considerable computing resources are required to quickly convert from one format to another.

Why do carriers need to transcode?
In all wireless applications, the radio channel used to communicate to the handset is a highly valuable resource. To maximize the efficient use of their channels, wireless carriers almost universally compress (or code) the speech using one of three standard algorithms: AMR, EVRC or QCELP. To maintain the voice quality, most wireless carriers prefer to keep the voice coded in one algorithm from one caller to the other.

Wireline carriers don’t have the same limitations on their extensive fiber optic backbone infrastructure, but they do have to squeeze one or more conversations over the customer’s DSL, Cable modem or wireless internet circuit. In much the same way the wireless carriers have adopted compression for the “last mile”, so have many of the wireline carriers. Because their needs are different (and to keep things interesting), the wireline carriers have chosen a different set of voice coding algorithms for their services: G.726, G.723.1 and G.729A.

As you can imagine, anywhere these different networks or applications need to connect to each other, there is a need for transcoding.

How to deploy transcoding?
Since transcoding is a conversion process and is fairly resource intensive, carriers generally will only transcode speech when absolutely necessary. Imagine that if two callers on the same wireline network are talking to each other and if both telephones use a common coding format (in this case G.729A), why bother transcoding? Most likely the call would go from one caller to the other without transcoding.

If instead, one of the callers needs to access an application server that only supports G.711, the conversion from G.729A to G.711 would be most efficient just before the application server. Typically, transcoding would be bundled with application servers, providing the conversion for the appropriate expected traffic.

What solutions does AudioCodes have?
AudioCodes offers a set of unique board-level solutions that can solve the transcoding needs of carriers and service providers:

IPM-260
In a PCI form-factor, the IPM-260 offers a highly efficient transcoding resource that can convert a wide list of wireline and wireless voice coders including: G.711, G.726, G.727, G.723.1, G.729A, AMR and EVRC. The IPM-260 is available in a range of densities from 30 ports up to 256 ports.
IPM-1610
In a cPCI form-factor the IPM-1610 provides a transcoding resource with similar capabilities of the PCI IPM-260, but with an increased density of up to 480 ports.
Trancoding Management – transcoding sessions can be established and controlled using either SIP or the AudioCodes VoIPlib API.

• SIP Control – Using optional SIP software, the transcoding control is accomplished via the Session Initiation Protocol (SIP), eliminating complex application development and operating system dependencies. Each leg of a transcoding session is established using an INVITE method, with the Uniform Resource Identifier (URI) for both calls uniquely tying the two sessions together.

• API Control – Using the standard API software, transcoding is accomplished using a set of API commands, establishing connections and controlling the voice coder selections.

Benefits of AudioCodes Transcoding Solutions:
• Available as embedded technology in a standard PCI or cPCI form factor
• Range of density solutions that can match the needs of a range of applications
• Wide range of simultaneous coder options
• Very low-latency and low-jitter
• Uses standard control protocols or proprietary API
• No drivers or operating system dependencies
• Meets the target cost per channel

Summary
As the market penetration of the VoIP carriers continues to grow, more systems will need to interoperate in pure IP integrations, many of which will require transcoding. AudioCodes has a number of unique solutions to help you solve integration requirements and helping you bridge the “transcoding gap”.