Richard wrote:

> Oh!  You mean a low cost, always on, broad-band wireless network?
>
> That would be 802.20 and THAT looks like the Intel Horse to back in
> the race... That's where the physical and medium access control layers
> would be specified and would make the whole idea of MIMO for the
> common man - work!

Ya know, I used to say the same thing, but the new "mobile profile" in
802.16 has essentially eclipsed 802.20.

Its politics, ya know.

Siavash Alamouti (Intel's "Broadband / Wireless" CTO) is (ahem) well
known in the MIMO community, having invented "Alamouti's technique"
which is the information-theoretic *most* efficient MIMO architecture. 
(Patented by ATT,but I digress.)

The new 802.20 is 802.16e.  (Approved and published as of Jan 31,
2006.)   Those of you following the story closely know that 802.16e was
originally scoped to use the existing 16a PHYs with minimal modification
adding MAC enhancements as necessary to support handover, and was to
maintain backwards compatibility with 802.16a (fixed wireless).

When the 16e PAR added the new PHY modes in 2003, the 802.20 group
protested to IEEE SA, and lost.  802.16e is a full-on cellular approach
(and potential replacement), and has Motorola and Qualcomm scared to death.

802.16e adds "Scalable OFDMA" (128, 512 and 1024 FFT) and expanded MIMO
support at the PHY layer, and a hybrid ARQ, handover support,
neighborhood advertisement, and a whole slew of security enhancements at
the MAC layer.

>
> (this is getting exciting!)

Indeed.  I may have to reverse my position on WiMax, at least as far as
16e is concerned.

Jim

> N6GPP
>
> On 2/21/06, *JIm Thompson* <[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>>
> wrote:
>
>
>     Mike Outmesguine wrote:
>
>     >Following last month's announcement <http://tinyurl.com/pbmoe> of
>     the first
>     >products to become WiMAX certified, the industry is all atwitter
>     with
>     >renewed excitement for the nascent technology. Though IEEE 802.16
>     is a few
>     >years old on paper, all the building blocks are just now coming
>     into place
>     >for Fixed WiMAX installations to be rolled out. Plus Wired News
>     reports
>     ><http://www.wired.com/news/columns/0,70241-0.html?tw=wn_index_2>
>     Mobile
>     >WiMAX could be upon us later this year.
>     >
>     >That said, it's time for So Cal Wireless Users to spend an evening
>     >discussing WiMAX and what possibilities it brings to our unique
>     mobile
>     >environment. Jack Unger will drive this discussion with a
>     presentation and
>     >Q&A session on the flavors of WiMAX and where it fits in to a
>     connected
>     >society.
>     >
>     >
>
>     "fixed" wireless, using WiMax or not, is, and will continue to be, a
>     limited market.
>
>     Quite frankly, you either operate WiMax in licensed spectrum (that you
>     have to 'buy") or you operate it in unlicensed spectrum, and have to
>     deal with the noise floor of a million other devices all using the
>     same
>     spectrum.  Using licensed spectrum is, essentially, uneconomic.  If
>     you're going to use unlicensed spectrum, you immediately run into
>
>     The WiMax PHY has no inherent advantage over the WiFi PHY, until you
>     decide to employ "smart antennas" or *potentially* MIMO, though on
>     the
>     MIMO front, its likely that the 802.11n effort will produce something
>     nearly as good, and much more available to the "common man".
>
>     The WiMax MAC has some advantages over the 802.11 MAC, but these
>     aren't
>     proprietary to 802.16.  I've long avocated a "MadMax" architecture,
>     using cheap, available 802.11 PHYs with an alternate, possibly quite
>     802.16-like MAC, using chipsets from Ralink, Atheros and others.
>
>     However, all that said, there is a 'mobile' profile in 802.16 now, and
>     that, *that*, my friends, is where the real WiMax money/action will
>     be.   Intel has plans (plus the will and money) to do no less than
>     erect
>     a "mobile Internet" service, ultimately replacing today's cell phones
>     with mobile computing devices (that will also do voice).   These will
>     all, of course, have some Intel chipset content inside.
>


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"SOCALWUG" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/SOCALWUG
SOCALWUG website: http://www.socalwug.org
-~----------~----~----~----~------~----~------~--~---

Reply via email to