Re: [codec] #17: Feedback and control loop?
"codec issue tracker" <trac@tools.ietf.org> Sun, 02 May 2010 18:08 UTC
Return-Path: <trac@tools.ietf.org>
X-Original-To: codec@core3.amsl.com
Delivered-To: codec@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 4F6EE3A68A7 for <codec@core3.amsl.com>; Sun, 2 May 2010 11:08:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.169
X-Spam-Level:
X-Spam-Status: No, score=-101.169 tagged_above=-999 required=5 tests=[AWL=-1.169, BAYES_50=0.001, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9t4XKyBtpMik for <codec@core3.amsl.com>; Sun, 2 May 2010 11:08:23 -0700 (PDT)
Received: from zinfandel.tools.ietf.org (unknown [IPv6:2001:1890:1112:1::2a]) by core3.amsl.com (Postfix) with ESMTP id 815FD3A6B1B for <codec@ietf.org>; Sun, 2 May 2010 11:08:17 -0700 (PDT)
Received: from localhost ([::1] helo=zinfandel.tools.ietf.org) by zinfandel.tools.ietf.org with esmtp (Exim 4.69) (envelope-from <trac@tools.ietf.org>) id 1O8daF-0003nv-CW; Sun, 02 May 2010 11:08:03 -0700
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: codec issue tracker <trac@tools.ietf.org>
X-Trac-Version: 0.11.6
Precedence: bulk
Auto-Submitted: auto-generated
X-Mailer: Trac 0.11.6, by Edgewall Software
To: hoene@uni-tuebingen.de
X-Trac-Project: codec
Date: Sun, 02 May 2010 18:08:03 -0000
X-URL: http://tools.ietf.org/codec/
X-Trac-Ticket-URL: http://trac.tools.ietf.org/wg/codec/trac/ticket/17#comment:2
Message-ID: <071.1ac44de1d442be78da8321bab758d21c@tools.ietf.org>
References: <062.4d06e5df10a25734771c5c65c2b40c5c@tools.ietf.org>
X-Trac-Ticket-ID: 17
In-Reply-To: <062.4d06e5df10a25734771c5c65c2b40c5c@tools.ietf.org>
X-SA-Exim-Connect-IP: ::1
X-SA-Exim-Rcpt-To: hoene@uni-tuebingen.de, codec@ietf.org
X-SA-Exim-Mail-From: trac@tools.ietf.org
X-SA-Exim-Scanned: No (on zinfandel.tools.ietf.org); SAEximRunCond expanded to false
Cc: codec@ietf.org
Subject: Re: [codec] #17: Feedback and control loop?
X-BeenThere: codec@ietf.org
X-Mailman-Version: 2.1.9
Reply-To: codec@ietf.org
List-Id: Codec WG <codec.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/codec>, <mailto:codec-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/codec>
List-Post: <mailto:codec@ietf.org>
List-Help: <mailto:codec-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/codec>, <mailto:codec-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 02 May 2010 18:08:26 -0000
#17: Feedback and control loop?
------------------------------------+---------------------------------------
Reporter: hoene@… | Owner:
Type: enhancement | Status: new
Priority: minor | Milestone:
Component: requirements | Version:
Severity: Active WG Document | Keywords:
------------------------------------+---------------------------------------
Comment(by hoene@…):
[Christian]:
Requirement: Reliable on the Internet
The ACS must be optimized towards acoustic real-time communications
over the Internet, and must have the flexibility to adjust to the
environment it operates in. Based on the quality of the end-to-end
speech packet transmission, the codec should adapt its quality and
delay to achieve an optimal benefit for the user.
As most Internet transport, it should be used with a wide range of
condition allowing a high reliability regardless the networking
condition. The reliability of the audio transmission should be high,
even in cases of low and varying bandwidth. This implies that the
codec is used on top of a transport protocol that implements a
congestion control algorithm and that the ACS adapts to changes of
available bandwidth. For example, if the available transmission
bandwidth is too low to allow the codec to transmit audio at a high
quality, the application can lower the sampling, bit or frame rate of
the stream at the cost of higher algorithmic delay or a degraded
audio quality.
Reiability and congestion control :
The acoustic transmission should be reliable and robust. The ACS
shall be not only robust against packet losses but also for periods
of low bandwidth.
The mean availability of the audio transmissions, calculated over all
users, might be one of the metrics for assessing the performance of
an Internet audio codec.
The ACS should adapt to the current network situation. Also, the
codecs of ACS themselves must be adaptable, because switching among
multiple codecs is difficult to negotiate and unlikely to work well
in situations of inter-operation.
Responding to congestion is a more complex issue and out of the scope
of this document. However, it shall be defined on how to use existing
congestion control protocols like DCCP and TCP. The ACS shall provide
the mechanisms that congestion control requires from the codec (i.e.
bitrate/framerate adaptability).
Because of the interactive nature of the acoustic transmission, the
bidirectional transmission of audio content can be used for
transmitting the required feedback and implementing a control loop.
Side channel
Congestion control should be must for all Internet applications also
for the ACS. [RFC3550] suggests in Chapter 10 somewhere that the RTP
profile should care for rate adaptation. Thus, the ACS should take
advantage of a feedback loop for variable coding parameter control in
order to allow a wide range of operation and to adapt to the the
current available bandwidth and processing power.
Congestion control per se is outside the review of this group, but
providing the hooks for a congestion-control mechanism to interact
with the codec is quite important. For example, running this codec on
a TFRC-enabled or DCCP RTP stream - TFRC and DCCP need to be able to
adjust (via the application) the bitrate of the codec in order to
implement congestion control and perhaps adjust packetization
periods/packet-rates.
A side channel for adaptation can be added. This would make sense
because in usage scenarios audio is always transmitted in both
directions. Adding a control channel would give a real advantage to
existing codec designs. Alternatively, such as side channel can be
also added with alternative solutions, such as handling that
communication in SIP/SDP and in RTP/RTCP.
--
Ticket URL: <http://trac.tools.ietf.org/wg/codec/trac/ticket/17#comment:2>
codec <http://tools.ietf.org/codec/>
- [codec] #17: Feedback and control loop? codec issue tracker
- Re: [codec] #17: Feedback and control loop? Colin Perkins
- Re: [codec] #17: Feedback and control loop? Cullen Jennings
- Re: [codec] #17: Feedback and control loop? codec issue tracker
- Re: [codec] #17: Feedback and control loop? codec issue tracker
- Re: [codec] #17: Feedback and control loop? codec issue tracker
- Re: [codec] #17: Feedback and control loop? Christian Hoene