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

Method and Apparatus for the Statistical Multiplexing of Voice, Data, and Image Signals

IP.com Disclosure Number: IPCOM000110400D
Original Publication Date: 1992-Nov-01
Included in the Prior Art Database: 2005-Mar-25
Document File: 3 page(s) / 171K

Publishing Venue

IBM

Related People

Irvin, DR: AUTHOR

Abstract

The purpose of the method described here is to control the time-averaged aggregate coding rate of a plurality of variable-rate digital source encoders by manipulating the encoders' individual short-term coding rates in such a way that the total number of frames transmitted over a packet network during a given interval of time is held to a pre-established budget, where that budget is set externally to achieve a desired financial objective and to provide an acceptable degree of coding fidelity.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 47% of the total text.

Method and Apparatus for the Statistical Multiplexing of Voice, Data, and Image Signals

       The purpose of the method described here is to control
the time-averaged aggregate coding rate of a plurality of
variable-rate digital source encoders by manipulating the encoders'
individual short-term coding rates in such a way that the total
number of frames transmitted over a packet network during a given
interval of time is held to a pre-established budget, where that
budget is set externally to achieve a desired financial objective and
to provide an acceptable degree of coding fidelity.

      This method is a conceptual enlargement of a prior technique
described in U.S. Patent 4455649, called here Esteban's method for
convenience.  Esteban's method itself extends in concept a yet older
technique known as Digital Speech Interpolation (DSI).  A DSI system
multiplexes N 64-Kbps digitized speech signals into M 64-Kbps
channels, where M is less than N, by removing the periods of silence
that naturally occur in human speech, and filling the resulting gaps
with segments of active speech from another source.  The method
proposed by Esteban allows digital data to be inserted into the gaps,
but, more importantly, provides for an adjustment in the coding rate
of the speech encoders to make room for more encoded signals once the
capacity of DSI to provide transmission compression is exhausted.
The goal for both a DSI system and for Esteban's method is, of
course, to minimize the need for dedicated transmission circuits.

      The system described here has the same general objective, but
works with a bandwidth-on-demand transmission service, rather than
with the fixed-capacity leased-line transmission circuits of the
earlier systems.  In the earlier systems, the bandwidth constraint
imposed by the capacity of the dedicated transmission circuits is
instantaneous and unforgiving; the instantaneous capacity of a T1
circuit, for example, cannot be exceeded under any conditions.  In
the system proposed here, however, the constraint specifies only the
number of packets transmitted over a bandwidth on demand transmission
system during a pre-established period of time; the constraint,
therefore, no longer imposes an inviolable upper limit on
instantaneous bandwidth.

      In overview, the method proposed here employs a global
management algorithm that is responsible for adjusting the rate of
each variable-rate source encoder under its domain.  This adjustment
is made in the least objectionable way that is consistent with
meeting a constraint on the number of packets transmitted during the
course of a pre-established time period.  When the management
algorithm expects transmission demand to exceed the pro-rated
constraint, it gradually decreases the short-term aggregate coding
rate.  When When the management algorithm expects transmission demand
to fall short of the pro-rated constraint, it gradually increases the
short-term coding rate...