Dear experts, Mr. Li,
we want to proceed with a technical and objective discussion.
Everybody is solicited to suggest further improvements for the provided test conditions to become as realistic as possible. Any constructive criticism is welcome as well.
Concerning your point Mr. Li, if you read the test conditions carefully you will see it is one item out of a list where you have the choice to take what you think is suitable. You will find the H.263 Wireless Profile also.
Furthermore we want to discuss the main points of the test conditions (bearer service, ...). What is your opinion about the UMTS scenario ?
We are still convinced, the provided test conditions are fair and reasonable.
Best Wishes Gero Baese
-----Ursprüngliche Nachricht----- Von: Adam Li [mailto:adamli@icsl.ucla.edu] Gesendet am: Montag, 16. Oktober 2000 07:00 An: Mailing list for parties associated with ITU-T Study Group 16 Cc: Baese Gero; Yung Lyul Lee; Jeong-Hoon Park; John D. Villasenor Betreff: RE: test conditions provided
Dear experts and Mr. Gero,
It seems that the conditions described in the document can hardly be called "reasonable and fair conditions".
For example, for Media codecs, it specifies H.263 with Annex O. As we all know, Annex O (with spatial, temporal and SNR scalability) is NOT in the Wireless Profiles of H.263. Indeed, it is not in ANY of the profiles defined in H.263.
The whole Annex O is such a very complex algorithm, that it is NOT even in the most complex profiles where complexity is not that much a concern. It is really unimaginable that any mobile applications will use Annex O.
If the Siemens proposal needed such a un-realistic scenario to provide any possible gain, the wireless connection oriented Annex I seems is not the right place for it.
Since Mr. Gero have just proposed the condition now, he may have enough resource to test it before the document submission date. Maybe you can test it also without Annex O, Mr. Gero?
Sincerely,
Adam Li
Adam H. Li, Ph.D. Image Communication Lab (310) 825-5178 (Lab) University of California, Los Angeles (310) 825-7928 (Fax)
-----Original Message----- From: Mailing list for parties associated with ITU-T Study Group 16 [mailto:ITU-SG16@mailbag.cps.intel.com]On Behalf Of Baese Gero Sent: Friday, October 13, 2000 7:38 AM To: ITU-SG16@mailbag.cps.intel.com Subject: test conditions provided
Dear experts,
the long lasting diskussion about the superiority of one of the proposals concerning error resilient transmission of progressive multimedia streams(H.323 Annex I) needs to be decided on an objective basis soon.
Therefor we are providing reasonable and fair test conditions.
The test conditions are focused on a common UMTS scenario with video input and PSNR as an objective quality measurement. Furthermore every step in between is well defined and a graphic visualization of the outcome is the ideal and easy basis, with not much room for interpretation, for the decision process.
Comments, questions and supplements are welcome.
Best wishes Gero Baese
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com
Dear experts and Mr. Gero,
Here are a few suggestions on the testing condition.
(1) Video codec
In Q15, there are two mobile configurations defined. They should be in the test conditions.
Profile 3 - H.263v2 Interactive and Streaming Wireless Profile Profile 4 - H.263v3 Interactive and Streaming Wireless Profile
The "simple scalable profile with Annex O" mentioned in the test condition is not one of the profiles defined in H.263. In addition, Annex O is not a commonly used effective option, especially in wireless application on a handset. It is suggested that this should be removed from test conditions.
(2) Lower layer testing condition
H.323 is a protocol over packet switched networks above the transport layers. It seems that for a simulation on the packet lose scenario to actually implement all the protocol stacks way down to channel coding of the data is unnecessary.
A reasonable assumption that the packetization is using RTP over an IP bearer with its characteristic error patterns should be sufficient. Indeed in your test condition documents, after all your analysis, you come to the conclusion that :
"Excessive link layer simulations with respect to the UMTS air interface have shown that a frame loss rate p in the range between 0.01 and 0.1 seems to be a reasonable assumption for current implementations of the chosen bearer service!"
This is exactly the impact that the H.323 will feel from all the lower layers. It is not necessary to regenerate this frame loss rate p by actually implementing all the protocol stacks.
Hope it helps.
Adam
---------- Adam H. Li, Ph.D. Image Communication Lab (310) 825-5178 (Lab) University of California, Los Angeles (310) 825-7928 (Fax)
-----Original Message----- From: Mailing list for parties associated with ITU-T Study Group 16 [mailto:ITU-SG16@mailbag.cps.intel.com]On Behalf Of Baese Gero Sent: Wednesday, October 18, 2000 5:54 AM To: ITU-SG16@mailbag.cps.intel.com Subject: AW: test conditions provided
Dear experts, Mr. Li,
we want to proceed with a technical and objective discussion.
Everybody is solicited to suggest further improvements for the provided test conditions to become as realistic as possible. Any constructive criticism is welcome as well.
Concerning your point Mr. Li, if you read the test conditions carefully you will see it is one item out of a list where you have the choice to take what you think is suitable. You will find the H.263 Wireless Profile also.
Furthermore we want to discuss the main points of the test conditions (bearer service, ...). What is your opinion about the UMTS scenario ?
We are still convinced, the provided test conditions are fair and reasonable.
Best Wishes Gero Baese
-----Ursprüngliche Nachricht----- Von: Adam Li [mailto:adamli@icsl.ucla.edu] Gesendet am: Montag, 16. Oktober 2000 07:00 An: Mailing list for parties associated with ITU-T Study Group 16 Cc: Baese Gero; Yung Lyul Lee; Jeong-Hoon Park; John D. Villasenor Betreff: RE: test conditions provided
Dear experts and Mr. Gero,
It seems that the conditions described in the document
can hardly be
called "reasonable and fair conditions".
For example, for Media codecs, it specifies H.263 with
Annex O. As we
all know, Annex O (with spatial, temporal and SNR
scalability) is NOT
in the Wireless Profiles of H.263. Indeed, it is not in ANY of the profiles defined in H.263.
The whole Annex O is such a very complex algorithm, that it is NOT even in the most complex profiles where complexity is not
that much a
concern. It is really unimaginable that any mobile
applications will
use Annex O.
If the Siemens proposal needed such a un-realistic
scenario to provide
any possible gain, the wireless connection oriented Annex
I seems is
not the right place for it.
Since Mr. Gero have just proposed the condition now, he may have enough resource to test it before the document submission
date. Maybe
you can test it also without Annex O, Mr. Gero?
Sincerely,
Adam Li
Adam H. Li, Ph.D. Image Communication Lab (310) 825-5178 (Lab) University of California, Los Angeles (310) 825-7928 (Fax)
-----Original Message----- From: Mailing list for parties associated with ITU-T
Study Group 16
[mailto:ITU-SG16@mailbag.cps.intel.com]On Behalf Of Baese Gero Sent: Friday, October 13, 2000 7:38 AM To: ITU-SG16@mailbag.cps.intel.com Subject: test conditions provided
Dear experts,
the long lasting diskussion about the superiority of one of the proposals concerning error resilient transmission of progressive multimedia streams(H.323 Annex I) needs to be decided on an objective basis soon.
Therefor we are providing reasonable and fair test conditions.
The test conditions are focused on a common UMTS scenario with video input and PSNR as an objective quality measurement. Furthermore every step in between is well defined and a graphic
visualization of
the outcome is the ideal and easy basis, with not much room for interpretation, for the decision process.
Comments, questions and supplements are welcome.
Best wishes Gero Baese
~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com
participants (2)
-
Adam Li -
Baese Gero