Voice Transcoding with IP Calls

A quick overview how Avaya Gatekeepers use the voice transcoding mechanism to convert voice transmission.

Voice Transcoding by Wellington Paez

Voice Transcoding with IP Calls

A quick overview how Avaya Gatekeepers use the voice transcoding mechanism to convert voice transmission.

As you read this post, “Voice Transcoding with IP Calls”, you will notice how this process takes place, which components become involve, and which ones are responsible for facilitating this process. You will also get to understand how this process is done on the Avaya Aura and IP Office systems.

When you start troubleshooting issues related to voice quality (QoS), you have to see how the customer has their network setup. Understand the VLAN, QoS rules, LLDP Settings, Codec types, among others. The recommendations for these settings change constantly. The manufacturer at times requires that you to let the Gatekeeper handle the call from cradle to grave. At other times, they suggest that you let the network devices control this process of Voice Transcoding. I wrote this post based on Avaya best practices and by the way, you are welcomed to disagree, just leave your explanation in the comment section.

The following steps help you understand how the Voice Transcoding works and its best practices.

  • 1.- What is Voice Transcoding
  • 2.- Hardware Types
  • 3.- What is shuffling, Hairpinning and Direct Media Path?

1.- What is Voice Transcoding 

Based on White Papers released August of 2009 by Avaya;  Voice transcoding is a voice signal converted from TDM to IP, or IP to TDM (with, or without compression and decompression). If calls are routed using multiple voice coders, as in the case of “call coverage” on an intermediary system back to a centralized voicemail system, the calls may experience multiple transcoding. Each transcoding episode results in some degradation of voice quality. Avaya helps reducing this process of Voice Transcoding degradation using Shuffling, Hairpinning, and Direct Media Path.

2.- Hardware Types

Depending which system you are implementing, there are different types of hardware needed to allow the Voice Transcoding to take effect. For Avaya Aura, there are some components to consider=

Port Network (PNs) Gateways (G650s, including the legacy MCCs, SCCs, Definity and Prologix)

A Port Network consist of a TDM (Time Division Multiplexing) and a Packet Bus, utilizing an IPSI (IP Server Interface) to connect to Ethernet Networks converting signaling messages to ethernet frames.

  • IP Media Processors (MedPro) – These Circuit Packs help to provision DSPs or VoIP resources to the PNs.
  • Control LANs (CLAN) – It can be configured as a Gatekeeper providing H.225 signaling to H.248 MGs and IP Phones.

H.248 Media Gateways (G450s and G430s, including the legacy G350s, G250s, and G700s)

They work very similar to the Port Networks, providing the same services, from registration functionalities, Voice Transcoding, etc.

  • Media Processors (MP80s and MP120s) – Just like the MedPros they provision DSPs or VoIP resources to the H.248 Media Gateways.

B5800s and IP Office – These are used for Distributed, Medium or Small deployments. Providing the same functionalities already mentioned above. They use VCM channels to do the Voice Transcoding.

  • Voice Compression Modules (VCM) – Just like MedPros and Media Processors, these boards provide DSPs or VoIP Resources with the same codec types (G711, 729, and 723). In some cases, Fax transmission with T.38 protocol might be also supported.

Gatekeepers work on lighting the IP Phones LEDs, Ring Control, Call Signaling and Registration and help with Failover rules.

Digital Signal Processors or DSPs work in encoding the audio with the different codec types (G711, G729, etc.). Using Voice Transcoding to do best effort in converting and matching the codecs across LANs.

3.- Shuffling, Hairpinning, and Direct Media Path

Avaya uses these three mechanisms to help the voice transcoding process by improving voice quality of these systems and using less of their resources. Avaya Systems use Shuffling, Hairpinning, and Direct Media Path to enhance system performance, helping the endpoints establish their connections and applying these rules once the call setup has taken place.

In order for these elements to work between point to point connections (IP Phone to IP Phone), the following has to be true=

  • H.323 or SIP has to be implemented
  • Support of the H.245 protocol
  • Codec Type

Now that we know how these mechanism work, lets take a look at them starting with=

Direct Media Path  The Avaya IP Office system uses this feature by selecting it in the IP Trunk or IP Extension form and it works by utilizing a Voice Compression Module resource for call setup, then it releases it as it keeps the call connected through the network.

Shuffling – IP Network Region Forms and Signaling groups is where you would configure this feature. It allows calls to be connected from an IP Phone to another IP endpoint using the same NR or Inter-NR, with similar Codec types. Once the call has been setup, the TDM Bus and time slots are released, utilizing the RTP Media streams to keep the call alive.

Hairpinning – Very similar to the shuffling mechanism. After the IP call gets setup between IP Phones, CM releases its TDM resources only leaving the MedPro or Media Processors channels connected throughout the duration of the phone call.

Question – Are you implementing these settings in your VoIP deployments?

Resources

The Signaling Group and IP-Network-Region forms contain the shuffling and hairpinning settings

Signaling Group Form IP Network Region Form

For IP Office Systems the Allow Direct Media Path are set under the Trunk and IP-Extension Forms, as seen here below

H.323 Extension Form IP Office H.323 Trunk Form

 

 

Avaya VoIP Quality Network Requirements – Page 13 

Please note: I reserve the right to delete comments that are offensive or off-topic.