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

RE: stds-80220-requirements: comments on rev5 - Channel bandwidth resolution



Title:
Joanne and Stewart,
 
Thank you both for your scholarly comments. May I also suggest that these comments be reflected in all the appropriate places in the System Requirements document, particularly sections 4.1.4 and 4.3 which are quite critical for the subsequent 802.20 standard development work. 
 
Dan 

 
 -----Original Message-----
From: Joanne Wilson [mailto:joanne@arraycomm.com]
Sent: Wednesday, August 20, 2003 8:40 PM
To: Wallace, Stewart J; Gal, Dan (Dan)
Cc: stds-80220-requirements@ieee.org
Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth resolution

Stewart,
 
I am in 100% agreement with all of your comments!
 
Best regards,
 
Joanne
-----Original Message-----
From: Wallace, Stewart J [mailto:Stewart.J.Wallace@team.telstra.com]
Sent: Wednesday, August 20, 2003 8:08 PM
To: Joanne Wilson; Gal, Dan (Dan)
Cc: stds-80220-requirements@ieee.org
Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth resolution

Joanne,
 
Yes - that's quite a good summary of the situation - I am directly involved in the ongoing ITU-R study groups, and also participate as a part of the Australian Delegation to the WRC's (incl the WRC-03 just finished).  Services (capital 'S') refers to ITU-R allocations - but, beneath that global structure, national administrations can make sub-allocations to "applications" (eg.within the MS we have: GSM, IS-95 CDMA, UMTS, cdma2000, etc.).  Furthermore, national administrations make "assignments" to specific operators/users (or classes of users) evidenced by the issue of "licences" to those operators/users.
 
Minor note - in Australia, we refer to the "unlicensed" applications as operating under a "Class Licence" which specifies generic technical characteristics, limits, etc. for operation in the relevant 'public park' spectrum allocations - eg. RLANs, cordless phones, garage door openers, model airplanes, ISM applications, etc..  Other spectrum bands are subject to explicit licensing of devices and/or operators.
 
Moving away from the detailed terminology, my original comments were simply aimed at ensuring that 802.20 did not unduly constrain itself as an emerging commercial technology by somewhat hastily mandating a 'wide' spectral occupancy granularity (> 1.25MHz) that had the effect of excluding it from many of the current national band structures.  As I mentioned in an earlier e-mail, national regulators are typically reticent (or at least extremely cautious) to redesign band plans, unless there is overwhelming need, since it is a complex regulatory task and it needs to be done carefully while bearing in mind lingering incumbent users who often cannot be readily displaced due to possible negative community impact.  See my earlier e-mail response to Dan Gal & the group.
 
One further point, while I made reference to the Australian 3.4 GHz band, this was only because someone earlier (~Mar'03?) suggested that it was a candidate band for 802.20 and I felt it useful to remind ourselves of the implications on spectral granularity of such a vision.  However, it seems to me that 802.20 is primarily a mobile application, and therefore more likely to be deployed in the MS allocations: eg. 800, 900, 1400/1500, 1700/1800/1900, 2100, 2500 MHz bands.  But, significantly, most of them are based on structures much more amenable to a 1.25MHz granularity, rather than a wider raster.  Thus, I strongly support a minimum granularity of 1.25MHz (but not larger) for 802.20.
 
In regard to the 3.4 GHz band, in actuality it may be that 802.16 is a stronger candidate for that particular band since it is based on a 3.5MHz channel plan - and the 3.4 GHz band is currently designated for Fixed Services (only - as opposed to MS) in Region 3, so we may confront certain global regulatory difficulties for 802.20 in this band, especially as it is also variously designated for Fixed Satellite Services.
 
In summary, I am suggesting that 802.20 be based on a minimum spectral granularity of 1.25MHz - and, yes, that may lead to some extra 'guard band', but at least we foster greater commercial deployment with minimal constraints.  Then, we leave it up to national administrations to decide if they wish to undertake band structure redesigns to squeeze a bit more utility out of particular domestic plans, by minimising unnecessary guard bands.
 
I hope this assists in moving toward consensus.

regards

Stewart J Wallace
Technical Regulatory Manager
Radiocommunications & Wireless Networks
Telstra Regulatory Directorate
Tel: (+61 3) 8627 8053
Fax: (+61 3) 9614 0670

-----Original Message-----
From: Joanne Wilson [mailto:joanne@arraycomm.com]
Sent: Thursday, 21 August 2003 7:31 AM
To: Gal, Dan (Dan); David Trinkwon; 'Michael Youssefmir'
Cc: 'Fujio Watanabe'; Klerer Mark; Wallace, Stewart J; stds-80220-requirements@ieee.org; Bharatula, Ganesh
Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth resolution

Dan,
 
It would be helpful if we were somewhat strict in our terminology.  At the international (i.e. ITU, WRC) level,
bands are allocated to different Services (capitalization here is important).  Those Services include the MOBILE
Service (MS), MOBILE SATELLITE Service (MSS), FIXED Service (FS), BROADCAST Service (BS), BROADCAST
SATELLITE Service (BSS), and others.    They are allocated on a PRIMARY or Secondary basis.  (Again, capitalization
here is important.)  None of this has anything to do with licenses.  This is specifically related to the allocations within the International Radio Regulations and such decisions are made at World Radiocommunications Conferences.
 
Administrations make national allocations within the context of the International Radio Regulations.  They can choose
to allocate (within their jurisdiction) MS spectrum for particular uses (i.e. services with a small "s"), including  Public Safety, Commercial Mobile Radio, etc., and determine (authorize) who gets to use that spectrum or how the spectrum is to be used .  Spectrum allocated (nationally) for exclusive use of a single party (operator) is LICENSED (capitalize here for emphasis only).  Those licenses are assigned by some mechanism -- the US uses auctions, other administrations use
that or other techniques. Or, as in the RLAN situation, they will allow equipment that meets a certain set of technical rules to share the band with no authorization (licensing) of the user required.  In the US this is referred to as an Unlicensed service and it is called License-Exempt in other markets.
 
I went through that primer to make the following point.
    1)  The PAR said "licensed" to mean systems that are operate in the band exclusively, not shared with other systems
         as in the RLAN case.  This was to distinguish 802.20 systems from RLANS and other systems that are Unlicensed
         and operate in a band on a shared basis.
    2)  The PAR didn't address whether the entity with the license to operate the 802.20 system was an "existing" or
         "new" operator.  Frankly, in the context of standards development, that's totally irrelevant.
    3)   The PAR said allocated to the Mobile Service because MBWA systems would meet the definition of a Mobile
          system and, from a regulatory perspective, and could only be operated within those allocations.  Identifying the MS
          allows one to identify blocks of spectrum where MBWA systems could be authorized to operate.  Specific bands are
          an issue for national regulators, not the ITU.
    4)   The PAR did not take a position on whether the spectrum had previously been allocated to the MS or not, because
          again that is irrelevant.
 
It's an interesting and, frankly, unconventional explanation of "existing" as meaning "allocated and available".  However,
under that definition, I don't know what you mean by when you say, "1.25 MHz and 5 MHz channels exist".  Did you mean that in some countries there is MS spectrum that is "allocated and available" for deploying a system with 1.25 MHz or 5 MHz channel bandwidth?  If that's what you meant, then I guess I would agree although I'm not sure what you mean by "available".  Regarding the problem of wasting .5 MHz of spectrum on internal guardbands, that's not at all unusual and is often required for two systems to operate effectively in adjacent bands -- even if they are both FDD systems.   I note that with 1.25 MHz/carrier, cdma2000 operators typically deploy only 3 carriers, not 4, within a 5 MHz allocation.  They are in fact choosing to have 625kHz guardbands on each side of their band.  So, using 2.5 MHz of spectrum for an MBWA system within a 3.5 MHz license and "wasting" 1 MHz of spectrum on internal guardbands fits with standard industry practice.
 
Best regards,
 
Joanne

 
 -----Original Message-----
From: Gal, Dan (Dan) [mailto:dgal@lucent.com]
Sent: Wednesday, August 20, 2003 2:39 PM
To: 'Joanne Wilson'; David Trinkwon; Gal, Dan (Dan); 'Michael Youssefmir'
Cc: 'Fujio Watanabe'; Klerer Mark; 'Wallace, Stewart J'; stds-80220-requirements@ieee.org; Bharatula, Ganesh
Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth resolution

Joanne & All,
 
Just to clarify:
In my previous message I wrote: "...the declared objectives of the PAR are also to deploy in existing licensed bands and to reuse existing infrastructure. "
The PAR does not say "existing" (vis-à-vis "Licensed spectrum").  I added this word to emphasize my interpretation of the PAR; that the IEEE 802.20 standard project would develop an air interface that is deployable in Licensed bands (i.e., bands allocated by regulatory bodies for Land Mobile communications service), not in some obscure future spectrum allocation. The word "existing" is not absolute - it is concurrent to the time the 802.20 standard is invoked. Thus, spectrum that does not "exist" today may very well "exist" (i.e., be "allocated and available") next year or ten years from today. The 1.25 MHz and 5 MHz channels exist today and are likely to exist in the next twenty years and beyond. Even if larger spectrum BLOCKS (e.g., 20 MHz, 25 MHz, or even 100 MHz) are allocated in the future for 3G/4G mobile communication service, a multi-carrier 802.20 solution could be deployed in such ultra wide frequency blocks. Thus, a choice of 1.25 MHz and 5 MHz channels is a future-proof.
 
In an earlier message, Stewart mentioned that in Australia, the 100 MHz block (3.425 GHz ~ 3.575 GHz) is divided  into 3.5 MHz  "units" (i.e., licenses) - for a TDD scheme I presume - and therefore he would like to see an 802.20 standard that specifies a channel (or channels) that would "fit" into one 3.5 MHz licensed spectrum. Obviously, a 5 MHz channel would not be suitable, but, is a 1.25 MHz channel acceptable? - after all, you would need to deploy a two-carrier solution (2x1.25 MHz, or one 2.5 MHz TDD channel) and waste the remaining 1 MHz  on two 0.5 MHz guard bands. Is my understanding correct?
 
 
Dan
 
 
 
 
 Thus, -----Original Message-----
From: Joanne Wilson [mailto:joanne@arraycomm.com]
Sent: Wednesday, August 20, 2003 9:57 AM
To: David Trinkwon; Gal, Dan (Dan); 'Michael Youssefmir'
Cc: 'Fujio Watanabe'; Klerer Mark; 'Wallace, Stewart J'; stds-80220-requirements@ieee.org; Bharatula, Ganesh
Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth resolution

All,
 
I agree with David that "existing bands" can include those he mentions below.  This argues for the
kind of 1.25 MHz granularity that had been proposed by Mike and others.  Of course, 802.20 systems
would be deployable whereever a service provider has sufficient spectrum and the regulatory authority
permits its deployment -- whether the service provider, the bands, and/or the infrastructure is "new" or "existing".
A key point raised in this discussion is how to strike the balance in granularity between providing enough
flexibility to allow 802.20 to be deployable in a variety of channel bandwidths while not dictating apriori a
particular technology or sub-channel structure.
 
Best regards,
 
Joanne
 -----Original Message-----
From: David Trinkwon [mailto:trinkwon@compuserve.com]
Sent: Wednesday, August 20, 2003 3:31 AM
To: Gal, Dan (Dan); 'Michael Youssefmir'
Cc: 'Fujio Watanabe'; Joanne Wilson; Klerer Mark; 'Wallace, Stewart J'; stds-80220-requirements@ieee.org; Bharatula, Ganesh
Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth resolution

"Existing Bands" does include MMDS in the US (and elsewhere) which are on a n x 6 MHz basis, and the 3.4 - 3.6 MHz bands in certain countries (various block / channelization schemes) - at least for "Portability" or "Nomadic" purposes if not full mobility. "Existing Infrastructure" obviously means "where applicable" and does not always imply "Existing Service Provider" especially when / where spectrum trading is introduced.

A major objective of 802.20 is to move the BWA industry forward beyond 3G, not just get stuck in the same ruts.


 
David Trinkwon
Email : Trinkwon@compuserve.com

USA Tel : 650 245 5650            Fax : 650 649 2728
UK   Tel : +44 (0)7802 538315  Fax : +44 (0)20 7504 3586
 


-----Original Message-----
From: owner-stds-80220-requirements@majordomo.ieee.org
[mailto:owner-stds-80220-requirements@majordomo.ieee.org]On Behalf Of
Gal, Dan (Dan)
Sent: 19 August 2003 21:06
To: 'Michael Youssefmir'; Gal, Dan (Dan)
Cc: 'Fujio Watanabe'; Joanne Wilson; Klerer Mark; 'Wallace, Stewart J';
stds-80220-requirements@ieee.org; Bharatula, Ganesh
Subject: RE: stds-80220-requirements: comments on rev5 - Channel
bandwidth resolution




Mike,

Well, yes. I recognize that the authors of the PAR did not want to commit the MBWA standard to 1.25 MHz and 5 MHz channels only (hence its language; "e.g., 1.25 MHz, 5 MHz"), yet, the declared objectives of the PAR are also to deploy in existing licensed bands and to reuse existing infrastructure. The above channels meet those criteria. There is little practical benefit in defining other ("future") channel-BWs in 802.20 unless such channels can be made available for deployment. Perhaps
it would be best if we leave this issue for work on future releases of IEEE 802.20?

Dan

-----Original Message-----
From: Michael Youssefmir [mailto:mike@arraycomm.com]
Sent: Saturday, August 16, 2003 12:02 PM
To: Gal, Dan (Dan)
Cc: 'Fujio Watanabe'; Joanne Wilson; Klerer Mark; 'Wallace, Stewart J';
stds-80220-requirements@ieee.org; Bharatula, Ganesh
Subject: Re: stds-80220-requirements: comments on rev5 - Channel
bandwidth resolution



Dan,

Just to be clear, I think the PAR only talks about 1.25MHz and 5MHz
as examples and not as mandates.

Mike


On Fri, Aug 15, 2003 at 03:47:38PM -0400, Gal, Dan (Dan) wrote:
>
> All,
>
> My view is that we should stick to the PAR definitions: 1.25 MHz and 5 MHz channel-bandwidths for FDD, and, if I understand correctly, 2.50 MHz and 10 MHz channels BWs for TDD. In future releases of IEEE 802.20, we may evaluate and adopt broader channels, as the evolving mobile wireless market may require.
>
> Dan
>
> -----Original Message-----
> From: Fujio Watanabe [mailto:fwatanabe@ieee.org]
> Sent: Friday, August 15, 2003 3:09 PM
> To: Joanne Wilson; Klerer Mark; 'Wallace, Stewart J';
> stds-80220-requirements@ieee.org
> Cc: Bharatula, Ganesh
> Subject: Re: stds-80220-requirements: comments on rev5 - Channel
> bandwidth resolution
>
>
>
>
> It is not practical to have one AI to fit a number of different bandwidths,
> although one may argue that SDR will enable it in the future. Since the
> technologies for an AI corresponding to a specified bandwidth (e.g., narrow
> band) are most likely different from those for another AI corresponding to
> another bandwidth (e.g., broadband), a system cannot be specified without a
> concrete value of bandwidth. For example, even if we tune some parameters of
> AI's in a PCS band, I don't think this AI can work in a broad bandwidth
> case, such as 100MHz required for the systems beyond IMT-2000 according to
> WRC'2003. Therefore, I would like to see several typical bandwidths
> specified for the MBWA.
>
> By the way, if 1.25MHz is called "broadband",  what will we call 100MHz?
> -- super broadband :)
>
> Fujio
>
>
> ----- Original Message -----
> From: "Joanne Wilson" <joanne@arraycomm.com>
> To: "Klerer Mark" <M.Klerer@flarion.com>; "'Wallace, Stewart J'"
> <Stewart.J.Wallace@team.telstra.com>; <stds-80220-requirements@ieee.org>
> Cc: "Bharatula, Ganesh" <Ganesh.Bharatula@team.telstra.com>
> Sent: Friday, August 15, 2003 11:42 AM
> Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth
> resolution
>
>
> >
> > I agree with Mark, Stewart, Arif and Mike on this point.  If we adopt
> > a plan for only 5, 10, 15,... MHz channel bandwidths we will limiting the
> > market opportunity for 802.20 systems unnecessarily.  From an economies
> > of scale perspective, I don't see how that would be in any of our
> interests.
> >
> > Best regards,
> >
> > Joanne
> >
> > -----Original Message-----
> > From: owner-stds-80220-requirements@majordomo.ieee.org
> > [mailto:owner-stds-80220-requirements@majordomo.ieee.org]On Behalf Of
> > Klerer Mark
> > Sent: Friday, August 15, 2003 10:00 AM
> > To: 'Wallace, Stewart J'; stds-80220-requirements@ieee.org
> > Cc: Bharatula, Ganesh
> > Subject: RE: stds-80220-requirements: comments on rev5 - Channel
> > bandwidth resolution
> >
> >
> >
> > I agree with Stewart and Arif. I believe we are trying to spec a system
> that
> > is deployable in the identified spectrum space and is scalable with
> existing
> > market demands and constraints.
> >
> > Mark Klerer
> >
> > -----Original Message-----
> > From: Wallace, Stewart J [mailto:Stewart.J.Wallace@team.telstra.com]
> > Sent: Thursday, August 14, 2003 10:59 PM
> > To: stds-80220-requirements@ieee.org
> > Cc: Bharatula, Ganesh
> > Subject: RE: stds-80220-requirements: comments on rev5 - Channel bandwidth
> > resolution
> >
> >
> > section 4.1.4
> >
> > In the case of Australia, I would just like to highlight that the 3.4GHz
> > band (covering 3.425 GHz ~ 3.575 GHz) has already been licenced under a
> > 15-year assured tenure regime, based on a lot size granularity of 3.5 MHz.
> > This approach was taken by our regulator in view of current FWA
> technologies
> > as the primary usage at that time.  I understand that there are several
> > other countries with similar band structures (although not necessarily
> with
> > the same tenure regime).  Thus, a 5MHz minimum channelisation restriction
> > would seem to potentially exclude Australia (at least) from the MBWA
> market
> > for the next 15+ years.
> >
> > In that context, I would suggest that a more flexible approach as
> suggested
> > by Arif would seem to be more prudent.
> >
> > regards
> >
> > Stewart J Wallace
> > Technical Regulatory Manager
> > Radiocommunications & Wireless Networks
> > Telstra Regulatory Directorate
> > Tel: (+61 3) 8627 8053
> > Fax: (+61 3) 9614 0670
> >
> >
> > -----Original Message-----
> > From: Ansari, Arif [mailto:Arif.Ansari@Nextel.com]
> > Sent: Friday, 15 August 2003 8:33 AM
> > To: Sheikh, Khurram P [GMG]; Fujio Watanabe;
> > stds-80220-requirements@ieee.org
> > Cc: Dennett, Steve
> > Subject: RE: stds-80220-requirements: comments on rev5
> >
> >
> >
> > A 1.25 Mhz channel bandwidth is consistent with the preferred North
> American
> > granularity and is the motivation for such a channel bandwidth in 3GPP2.
> > The original text included 1.25 and 5 MHz as examples, again consistent
> with
> > other standardization efforts to not make the channel bandwidth a
> > requirement.  This adequately covers the mobile licensed band worldwide,
> and
> > the follow-on text also included the possbility of wider channels.  At the
> > minimum, I would suggest that 1.25 MHz not be excluded, while 5 MHz and
> > multiples thereof can also be included.  Ideally I would like to see all
> > these channel bandwidths as no more than examples.
> >
> > -----Original Message-----
> > From: Sheikh, Khurram P [GMG] [mailto:khurram.p.sheikh@mail.sprint.com]
> > Sent: Thursday, August 14, 2003 4:19 PM
> > To: Fujio Watanabe; stds-80220-requirements@ieee.org
> > Subject: RE: stds-80220-requirements: comments on rev5
> >
> >
> >
> > I would like to add to Fujio's comments and my earlier contribution.
> > Multiples of 5 MHz is critical for both a technical performance as well
> > economic viability (capital efficiency) given other performance
> > parameters (system throughput, number of users, broadband data models
> > etc.)
> >
> > Thanks and look forward to any rationales why less than 5 MHz could be
> > an option for the MBWA system tied to our current performance
> > requirements.
> >
> >
> >
> > Khurram P. Sheikh
> > Chief Technology Advisor
> > Sprint- Broadband Wireless
> > Tel (SM): 650-513-2056
> > Tel(KC): 913-762-1645
> > Mobile: 650-906-8989
> > khurram.p.sheikh@mail.sprint.com
> >
> > -----Original Message-----
> > From: Fujio Watanabe [mailto:fwatanabe@ieee.org]
> > Sent: Wednesday, August 13, 2003 8:57 PM
> > To: stds-80220-requirements@ieee.org
> > Subject: RE: stds-80220-requirements: comments on rev5
> >
> >
> >
> > I would like to make a comment on John's email of July 23rd on section
> > 4.1.4
> > as follows.
> >
> > I don't agree to eliminate this section (John said "stricken") because
> > the
> > bandwidth is one of important basic system requirements.  A system
> > cannot be
> > specified without concrete values of bandwidth.
> > A broader bandwidth is a current trend of wireless communications, e.g.,
> > WLAN (e.g., 20MHz), UWB (e.g., >300MHz), possible systems beyond
> > IMT-2000
> > (e.g., 100MHz) as well as a general requirement for Mobile "Broadband"
> > Wireless Access.
> >
> > I also understand John's rationale to not limit the lower bound of the
> > bandwidth.
> >
> > Therefore, how about to have several typical numbers for the bandwidth
> > as
> > options in this section?
> >
> > Best Regards,
> >
> > Fujio
> >
> >
> > > -----Original Message-----
> > > From: Fan John [mailto:J.Fan@flarion.com]
> > > Sent: Wednesday, July 23, 2003 6:15 PM
> > > To: 'stds-80220-requirements@ieee.org'
> > > Subject: stds-80220-requirements: comments on rev5
> > >
> > >
> > >
> > > Hi all,
> > >
> > > These are comments on rev5 of the document from Marc
> > > Goldberg, Michael Youssefmir, Samir Kapoor, Joanne Wilson,
> > > Arif Ansari and John Fan.
> > >
> > > --John
> >
> > > 4.1.4. Channel Bandwidth
> > >
> > > Action: This section should be stricken.
> > >
> > > Rationale: The current text requires "multiples of 5 MHz" for
> > > deployment. No rationale for 5Mhz has been given on the
> > > reflector.  Beyond that, a 5 MHz minimum bandwidth would
> > > limit the applicability of the MBWA AI in many of the
> > > available licensed bands below 3.5 GHz.
> > >
> >
> >
> >
> >