Re: [STDS-802-16] Unicast Polling 802.16d/D5 question
To answer David Putzolu's original question, there is no correction needed in Section 6.3.6.3.1 regarding the padding of un-used bandwidth request opportunity provided by unicast. This is because the padding mechanism described in Section 6.3.3.7 does not limited to partially used UL data bursts, it actually applies to a fully unused UL allocation, such as a bandwidth request opportunity by unicast polling. It belongs to the case of the second sentence of Section 6.3.3.7, i.e., MAC PDU with padding CID.
There were discussions regarding the padding issue in previous 802.16 meetings. Ken's response explains the intention of having SS to send padding even if it has nothing to send. Basically it provides the BS a NACK (Negative ACKnowlegement). However, there was a debate regarding whether or not to pad to occupy the entire UL allocation, for power saving and noise reduction, etc. Personally, I think those issues are still valid for further discussions, particularly for TGe.
Well, after all, a good implementation shall minimize any potential paddings, particularly, the use of MAC PDUs with the padding CID. If turns to be very bad, then it either have allocation granularity issues or the scheduler does not do a good job. --:)
Lei
-----Original Message-----
From: Johnston, Dj [mailto:dj.johnston@INTEL.COM]
Sent: September 20, 2004 6:19 PM
To: STDS-802-16@listserv.ieee.org
Subject: Re: [STDS-802-16] Unicast Polling 802.16d/D5 question
I believe the issue is with mobiles, where the power expense of
transmitting unnecessarily might not be a good idea. Does .16e have a
get out?
DJ
-----Original Message-----
From: owner-stds-802-16@listserv.ieee.org
[mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Kenneth
Stanwood
Sent: Monday, September 20, 2004 5:16 PM
To: STDS-802-16@listserv.ieee.org
Subject: Re: [STDS-802-16] Unicast Polling 802.16d/D5 question
Unless it has changed for specific PHYs, the original intent was to have
an SS transmit padding if it was given an allocation when it had nothing
to send. Otherwise, it may not be clear to the BS whether the
transmission was intentionally missing our just very errored.
Ken
-----Original Message-----
From: owner-stds-802-16@listserv.ieee.org
[mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Putzolu, David
Sent: Monday, September 20, 2004 3:45 PM
To: STDS-802-16@listserv.ieee.org
Subject: [STDS-802-16] Unicast Polling 802.16d/D5 question
I am reading the 802.16d/D5 spec & have found text in the unicast
polling section which is a bit surprising.
Section 6.3.6.3.1, which describes BS use of unicast polling to query
individual SS for bandwidth requests, states, "If the SS does not need
bandwidth, the allocation is padded in accordance with 6.3.3.7"
Section 6.3.3.7 describes how padding is done for "allocated space
within a data burst that is unused."
Reading these two sections gives the impression that if a SS receives a
grant allocation and has nothing to transmit, it must transmit padding
anyway.
Is this reading of 6.3.6.3.1 correct? Is there some other part of the
spec that clarifies this?
Thanks,
David