Interesting discussions from the bowels of 3GPP
----------
For background, HSDPA is 3GPP's version of 1xevdv - High speed data + voice.... presently under development for probable inclusion in UMTS specification release 5.
They are considering a "standalone DSCH for HSDPA" - which appears to be a data-only option similar to 1xevdo
It appears that there were conversations considering an OFDM based solution for the physical layer for "standalone DSCH" - as well as the wCDMA based physical layer planned for their 1xevdv version.
After discussions, only the wCDMA based solution was retained for the "Standalone DSCH" related recommendation.
Here are email conversations relating to the issue:
----------
----------
Subject:
Re: Draft TSG RAN WG1 report for TSG RAN#11
Date:
Fri, 9 Mar 2001 15:46:32 -0000
From:
Sarah Boumendil <boumendi@NORTELNETWORKS.COM>
To:
3GPP_TSG_RAN_WG1@LIST.ETSI.FR
Dear Antti,
thanks for the report. Regarding the standalone DSCH for HSDPA, I think we had agreed during the meeting that the OFDM based solution would not be mentioned in the TR however it would be mentioned in your report to the RAN to give some visibility to the other working groups on the discussions which took place in RAN1 regarding possible alternative solutions for HSDPA.
Could you add it in your report? I guess something like "An alternative proposal for HSDPA called standalone DSCH has been discussed in RAN1#19 : HS-DSCH is mapped on a downlink carrier that is different from the carrier that supports its associated DCH/DPCH. 2 solutions have been discussed a WB-CDMA based and an OFDM based, only the WB-CDMA based was retained for the TR" would be ok.
Thanks in advance,
Sarah
----------
Subject:
Re: Draft TSG RAN WG1 report for TSG RAN#11
Date:
Fri, 9 Mar 2001 17:39:36 +0100
From:
"Erik Dahlman (ERA)" <Erik.Dahlman@ERA.ERICSSON.SE>
To:
3GPP_TSG_RAN_WG1@LIST.ETSI.FR
Dear all,
I agree with Sarah that the "Stand-alone DSCH" discussion should be mentioned in the
report. However:
- I do not think the "Stand-alone DSCH" based on WCDMA was discussed is an
"alternative solution". Rather, it was presented as a more or less straightforward
generalization of what had up until now been discussed within the HSDPA study item
(a "super-set"). With straighforward I mean straightforward from a spec. point-of-view
but obviously not from a UE implementation point-of-view :-) .
- In general, I am somewhat concerned about the liberal use of terminology. To me "Stand-alone
DSCH" indicates the already very well specified DSCH modified so that it can be mapped to a
separate carrier. In the same way HS-DSCH is defined in the TR(s) and the TR(s) describe some
basic principles of the physical layer structure used for the HS-DSCH (HSDPA) transmission.
It is then rather confusing when, in the "Stand-alone DSCH" discussion, DSCH and HS-DSCH
may suddenly mean something completely different.
Thus I would prefer if the "Stand-alone DSCH "discussion in RAN1 could be described somewhat
like this.
- The possibility for supporting Stand-alone DSCH/HS-DSCH mapped to a carrier that is different from the
carrier that supports its associated DCH/DPCH was discussed during RAN#19.
{ I do not mind if someone, at this stage, want to mention the RAN1 conclusions, e.g. related to
multi-carrier cells}
- Related to this, there was also a discussion on using a different physical layer based on OFDM for the
stand-alone carrier.
I agree with Sarah that it should be emphasized that the RAN1 "Stand-alone-DSCH"-related
recommendation in the RAN1 HSDPA TR relates to the WCDMA-based solution.
By the way, we talk about WCDMA and OFDM based solutions. I guess stand-alone DSCH is eqully
applicable to TDD so we should probably talk about UTRA-based vs. OFDM-based solutions or
something like this.
Have a nice week-end.
ED
----------
----------
ERICY - and other major members of the GSM guild - are also members of the OFDM Forum - one of 2 competing OFDM organizations focused on the standardization of the technology. The other, the BWIF, is founded by broadband types like CSCO. There is no cross membership - the 2 organizations are entirely polarized, with legal disputes erupting between respective members.
Within the OFDM Forum, where telecom vendors reside, no discernable progress has been made within the Broadband Mobile Wireless Working Group, nor are any significant vendor members participating. They are, however, participating with fixed wireless and WAN development.
So - provided the "Standalone DSCH" in question is mobile wireless, the absence of progress within the OFDM Forum would explain its rapid dismissal.
The OFDM company who has announced a mobile wireless solution - Flarion, with its 384kbs demonstration - does not belong to either of the competing OFDM Forums. As a result, the GSM vendor guild within 3GPP, who are also members of the OFDM Forum, would not consider Flarion technology or standards "valid" since they've not been subject to committee based "development" within the OFDM Forum.
Flarion is developing into a rouge elephant - much like Q - and will have to commercialize its technology on its own, and against the tide.
The GSM vendor guild would, no doubt, like to evolve toward a non-Q technology like OFDM (although I understand that HDR is, in part, based on OFDM technology). But fleeing Q's benevolent grip for Flarion's would not be a palatable prospect.
Will it be possible for the OFDM Forum to develop a broadband mobile data solution while working around Flarion's technology? The challenge sounds all to familiar. They tried working around Q's cdma - and failed - with wCDMA. Also, it's worth noting that the promoted efficiencies of OFDM over cdma are being questioned. And Flarion's 384kbs demonstration speed is not impressive when contrasted with 1xevdo's performance.
Now that the GSM guild is entrenched in wCDMA/UMTS - having diluted the standard's IP and having worked its way into 3GPP2 1xev standards development - how anxious will it be to roll the dice in hopes of securing better control over another technology?
It's interesting to note that the 3GPP perceives need for a data-only option. Is this in response to a perceived attraction for 1xevdo by GSM - and particularly European GSM carriers? The 3GPP's data-only specification will, however, arrive well late of 1xevdo's receipt of imt2000 certification... leaving GSM guild vendors in the awkward position of asking carriers to wait... again.
ben |
