Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Coordinated Compressed Connection

IP.com Disclosure Number: IPCOM000116710D
Original Publication Date: 1995-Oct-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 4 page(s) / 96K

Publishing Venue

IBM

Related People

Galand, C: AUTHOR [+5]

Abstract

Voice networks typically consist of a group of switching elements interconnected by communication links. The switching elements can include Private Branch Exchanges (PBXs), Central Office Switches (COS), and/or Tandem Switches (TS)). The links are typically T1 or E1 digital circuits operating at 1.544Mbps or 2.048Mbps, respectively.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Coordinated Compressed Connection

      Voice networks typically consist of a group of switching
elements interconnected by communication links.  The switching
elements can include Private Branch Exchanges (PBXs), Central Office
Switches (COS), and/or Tandem Switches (TS)).  The links are
typically T1 or E1 digital circuits operating at 1.544Mbps or
2.048Mbps, respectively.

      In order to reduce the transmission costs, voice compression
devices (or Voice Servers (VS)) are deployed in pairs on the digital
links interconnecting the switching elements.  On the links where a
pair of Voice Servers are used, the bandwidth needed to transport
voice traffic can be reduced to less than 20% of the previously
required, uncompressed bandwidth.  Thus, the deployment of
compression devices can result in a very attractive and substantial
financial savings for the service provider.

      However, when several compression/decompression cycles are
cascaded on a voice signal, the delays and distortions are
accumulated, and may become annoying.  The following method can be
used to eliminate this problem and ensure a single
compression/decompression cycle on a telephone call.  For the sake of
simplicity, consider 13 kbps GSM coders, but the method can support
multiple coding techniques.

      Every GSM decoder characterizes the 64 kbps PCM voice signal
that has been decoded from a 13 kbps bit stream in such a way that
whenever this signal is forwarded to another compressor for the next
hop encoding, this GSM compressor detects that the signal has been
decoded from a GSM coded speech.  If this compressor supports GSM
coding, then it signals back to the decoder that it is ready to
receive directly the coded speech instead of the decoded speech.

      The characterization of the 64 kbps stream can be seen like an
in-band signaling between decoder (e.g., VS_2) and coder (e.g.,

VS_3), which must be such that whenever the PCM voice signal is not
looped back at the tandem switch (e.g., TS_1), but directly sent to a
user, there is no audible degradation in the decoded voice.

At call set-up, the Voice Server in-band signaling includes two
phases:
  1.  Hand-shaking phase:
   ...