| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
Dear Yossi and Hongmei,
In p.498 of P802.16-Revd/D5, there is a definition of PUSC slot :
"For downlink PUSC using the distributed subcarrier permutation (defined in 8.4.6.1.2.1),
one slot is one subchannel by two OFDMA symbols."
According to the above definition, the number of data subcarriers per subchannel is 24,
and the number of data subcarriers per slot is 48 for PUSC.
Best Regards,
Seung Joo
------- Original Message -------
Sender : Yossi
Segal<yossis@RUNCOM.CO.IL>
Date : 2004-11-04
19:13
Title : Re: [STDS-802-16] Regarding Band AMC--Mapping
|
Hi Hongemi,
In tables 308b/c/d there is a mistake in the description, it should have been as in table 308 in the 802.16d: " Number of data subcarriers in each symbol per subchannel" which has no ambiguity.
48 subcarriers in a sunchannel, but for PUSC there are 24 subcarriers allocated per OFDMA symbol (over 2 symbols gives you 48).
Regards, Yossi From: owner-stds-802-16@listserv.ieee.org [mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Sun, Hongmei Sent: Thursday, November 04, 2004 11:43 AM To: STDS-802-16@listserv.ieee.org Subject: Re: [STDS-802-16] Regarding Band AMC--Mapping Dear Panyuh Joo:
Thank you so much for you reply. For AMC, I think I'm quite clear now. For DL PUSC, I think I can understand your meaning, but I do think description of Spec is confusing.
For example, in Table 308b, Table 308c and Table 308d of 802.16e/D4 (page 145~146), there is one row: "Number of carriers per subchannel: 24 " So doesn't that mean in DL PUSC, the each subchannel is composed of 24 subcarrier ?
But in the other hand, in DL PUSC, 2 cluster makes 1 subchannel, but 1 cluster=12 subcarrier X 2 sym, right, so from this point, 1 subchannel has 48 subcarrier totally. No matter what this makes reader confusing at its first reading.
I think the spec doesn't not explicitly seperate these 2 cases, whether the subcarrier is counted focusing on freq. domain only, or countered by taking both time domain and freq. domain altogether.
Hongmei SUN
From: owner-stds-802-16@listserv.ieee.org [mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Panyuh Joo Sent: Wednesday, November 03, 2004 4:15 PM To: STDS-802-16@listserv.ieee.org Subject: Re: [STDS-802-16] Regarding Band AMC--Mapping Dear Hongmei
Sun, The reason of the following description exist is
for easier understanding the concept to the
band. “AMC subchannel consists of 6
contiguous bins in a same band." And because the statement does not restrict the
use of 6 contiguous bins within one symbol, one may use 4 bin in first symbol and the other 2 bin in the second
symbol. If someone uses the abbreviation on this structure as “ Maybe the word “contiguous” give some wrong sense
of instincts. If just forget the nuance of the word “contiguous” as
continuous in frequency domain, just read it as Contiguous bin in Band.
Then we may bend the thoughts as well as the
words. And also please just do not confuse by the concept
of the Band. It just gives you the concept of the AMC subchannelization. And it is not used for BAND AMC
MAPPIING procedure by itself. The mapping of the data PDU is aligned only
with definition of subchannel. I still do not see any mis-description in the original statement in the
specs. And regarding to the DL PUSC, I think there are
some mis-understanding when you read the spec.,
it would be appreciated if you can revisit the spec and read it carefully.
I will do it too. If I found the same problems, I will comeback with the
answer. But currently, I do not found any fault in the
spec. Hope this may help U. Panyuh Joo From:
owner-stds-802-16@listserv.ieee.org
[mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Sun, Hongmei Panyuh, sorry, but I'm
still confused about this even after your
explanation. As you said, 4
Bins make one Band, and 6 Bins make one subchannel. So doesn't this
hint that each AMC subchannel must exist across Band? (My
understanding is: for each subchannel: 4 bins are from 1st Band, the
other 2 bins are from 2nd Band. Of course, these 6 bins are
contiguous.) In addition,from
my understanding, maybe PUSC DL permutation scheme is an exception
case in aspect of subchannel size, where each subchannel has 24 tones
but not 48 tones. Pls feel free to
revise me if my understanding is not right. Hongmei
SUN CTL,
Intel From:
owner-stds-802-16@listserv.ieee.org
[mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Panyuh Joo Hi,
Varma and All, Because
one Band only got 4bins and 4 Bin only can make 32 data subcarriers + 4
pilot subcarriers. Any subchannel shoud have at least 48
tones(subcarriers), if we do not restrict the definition of AMC
subchannel into one band. than any AMC subchannel can be exist
accross the Band. So, the
meaning of "AMC subchannel consists
of 6 contiguous bins in a same band"
==> one AMC subchannel should be confined with in the same band. This
definition would help the system exploits the Multi user diversity.
I hope this explanation would help
you.
|