Re: [802-20-GENERAL:] Letter Ballot 1 Proposed Comment Resolutions Poste

Thread Links	Date Links
Thread Prev	Thread Next	Thread Index	Date Prev	Date Next	Date Index

I've partially read the comment resolutions and the revised draft, and I found some resolution is not so desirable and have room to prove. So I give some suggestions as following. And I have to state that the following reply are just partial of my response of the resolution, and I'd like to preserve my right to submit more response suggestion for resolution in future. BTW, I suggest to postpone the deadline of collecting response, for the 4 working day is really not enough for the WG member to review and provide response.

1, Comment No. 10 was not completely resolved. The resolution was "Accepted as modified: See resolution of comment 63".

The comment 63 is only focus on the different format of overview, while the original comment required to provide unified abbreviations and describe the two modes using the same format, etc. And these requirement is reasonable, so I don't think the resolution committee resolved this comment completely.

2, Comment No. 26 was responded as "Non-actionable comment/question" for the reason of "The requirements are not meant to be met simultaneously under all worst case conditions".

While in the comment, only the mobility requirement and spectrum requirement in PAR be cited simultaneous, that is the WG objective performance which should not be regard as the worst case condition. So, as the commenter, I ask for a clarification to answer this issue.

3, Comment No. 144 was about the copyright issue, the resolution reply is "Copyright clearance or a "fair use" statement for specific sections that have been included in the draft has been requested in writing from the organizations owning the copyright. Responses are anticipated prior to final recirculation."

The remedy seems not appropriate, because all the comment should be resolution before recirculation defined in 802.20 P&P. So I suggested to change that "Copyright issue claearance" or a "fair use" should be achieved before next recirculation.

4, Comment No. 399 is of procedural/requirements nature:The draft standard do not address Radio Transmitter and Receiver Requirement subject of Sections 4.2.5.2 and 4.2.5.3 of 802.20 SRD which is a requirement for compliance and completeness as defined in 802.20 TSP.

I believe this is appropriate, and the transmitter and receiver part is also provided by other standard such as 802.16.

5, Comment No. 492: The OFDM symbols in the RL are extended, windowed, and adjacent symbols have some overlap. This is a problem since different frames may be transmitted by different ATs, and the overlap of signals between ATs at the frame boundary could cause significant intra-cell interference if either 1) the timing advance value of an AT is not perfect (which is likely), or 2) there is a large power difference between the signals from different ATs as received at the AN.
While the resolution committee stated that: "Declined: The adjacent symbols in different frames do have some overlap, however this overlap is only during the windowed portions that are not used for demodulation."

That is based on assumption that subsequent OFDM symbols overlap no more that windowed period, would you please let the drafter to provide verification for that?
And the explanation "The timing alignment between different terminals has to be within a cyclic prefix duration in order for them not to collide with each other."
I do not think this can be always guaranteed especially when multipath delay and RL timing errors are present.
"Moreover, the draft spec contains a CDMA control segment that allows for very accurate timing estimation. "
How accurate the timing estimation is? And cyclic prefix must compensate for multipath delay, synchronization errors, RL timing errors and windowing period. Multipath delay is variable and adds on top of other delays.

6, Comment No.20 :DiversityHoppingModeOFF is suitable for low speed mobile since this makes it possible to exploit multiuser diversity in frequency domain. On the other hand, we cannot exploit multiuser diversity in frequency domain for high speed mobile due to rapid channel variation. For high speed mobile, it is beneficial to use DiversityHoppingModeON to increase link reliability. MBFDD/TDD can use only one mode in a carrier. In case of MulticarrierOFF mode, therefore, there should be only one mode in the system. If the system uses DiversityHoppingModeON, there are limits for increasing the capacity of the system because we cannot exploit multiuser diversity in frequency domain. In case of DiversityHoppingModeOFF, high-speed mobiles have unreliable links and hence the performance can be degraded. While the resolution committee stated that: "Declined: There is no value in enabling both channelization modes simultaneously. The main advantage of DiversityHoppingModeON is low overhead of common pilot while DiversityHoppingModeOFF allows for performance gains through e.g. sub-band scheduling, interference estimation per tile (tile = 16 contiguous tones over PHY Frame), as well as efficient channel estimation with precoding/beamforming, SDMA and power control. The choice between DiversityHoppingModeOFF and DiversityHoppingModeON depends on the availability of performance gains achievable with DiversityHoppingModeOFF which is deployment specific (e.g. synchronous versus asynchronous, use of advanced multi-antenna techniques). Hence, there is no value in enabling both channelization modes simultaneously."

But It can not be assumed that system will handle only low mobility or high mobility users. In typical deployment scenario a mix of users will be present hence one group will suffer defending on the DiversityHoppingMode which is a system parameter. Ideally both modes shall be supported.

7, Comment No. 238 :The number of FL sub-bands is too small for getting full gains from frequency-selective scheduling. Namely, the only supported sub-band bandwidths are approximately 1.25 MHz and 2.5 MHz. While the resolution committee stated that: "Declined: The sub-band sizes given in the specification are appropriate" because "Interference diversity diminishes, so for smaller subbands, a user on one sub-band will see interference from only one interferer, and adverse effect when this interference is high"(the reason 3).

But this is not quite true if smaller bands suffer interference that means that this band shall not be uses!!!!. Very narrow band interference would be still mitigated by frequency selective scheduling.

And generally, some comments were not accepted on the ground of simulation results, so some analysis is necessary to verify the simulation results.

Best Regards,
**************************************************

Lin, Jiezhen

Corporate Technology
Mobile Network Standardization

Siemens Ltd China
Tel: +86 10 6476 6914
Fax: +86 10 6475 9216

GSM: +86 135 0108 1309
Email: jiezhen.lin@siemens.com

From: Jerry1upton@AOL.COM [mailto:Jerry1upton@AOL.COM]
Sent: 2006年4月29日 5:44
To: STDS-802-MOBILITY@listserv.ieee.org
Subject: [802-20-GENERAL:] Letter Ballot 1 Proposed Comment Resolutions Posted

The Letter Ballot 1 proposed Comment Resolutions are posted in the Members only section of the website. Also a revised Draft is posted based on the proposed Comment Resolutions.

The Comment Resolution Committee members reviewed all posted resolutions. The members of the committee are:

All members of 802.20 have access to the comments, comment resolutions and the updated drafts regardless of their voting status for Letter Ballot 1. All Members may send their suggested comment resolution improvements to the Editor, Mark Klerer, and Chair. All suggested improvements are requested by close of business on May 4, 2006.

Additionally, all No voters in Letter Ballot 1 will be contacted individually to address the resolutions of their negative ballot comments.

Re: [802-20-GENERAL:] Letter Ballot 1 Proposed Comment Resolutions Posted