Final nail in Mobile WiMax coffin?

Michael Watterson's picture

Fixed WiMax has a sure future, but Intel's dream of bundling their chips for Mobile WiMax into every Notebook, Laptop, Netbook and Tablet is fading. There were only two major Mobile Operators that adopted Mobile WiMax. The Russian operator is already switching to LTE for their roll out and Clearwire (USA major cities) appear to have a "road map" to switch to LTE in next couple of years.

Imagine's Irish Wimax using Intel and Motorola gear with Intel backing is giving WiMax a bad name as building penetration at 3.6GHz is very poor, though an improvement over the S-CDMA based Ripwave that Imagine "Inherited" on their 3.6Ghz licences when they bought IBB from NTR.. Clearwire in Ireland had a very poor reputation and the operation has been sold to Imagine.


Citing the ever-present "people familiar with the situation", the Wall Street Journal
reckons that Intel will have to pay around €1.5bn for the unit, which
makes baseband and platform chips for GSM-family handsets (GSM/WCDMA).
Such a deal would allow Intel to provide manufacturers with a complete
chipset, in the way that Qualcomm already does.

Infineon has a LTE chip(set) as well as the older Mobile Wireless chippery.

While this helps Intel's ambition for Meego and Android handsets, Smart Phones, tablets or MIDs running Atom, it also lets them leverage the Centrino Brand for LTE and 3G/HSPA+ as well as WiFi in Notebooks, Laptops and Netbooks.

Will Imagine have been the last major backing of Mobile WiMax technology rollout? Conversely WiMax is very successful in the Fixed Wireless Broadband market, that always uses outdoor aerials and above 2GHz usually the entire "radio" and aerial array is on one neat outdoor box. The only serious competitor in that market is the DOCSIS over Wireless and related "WiDox" from Arris (which is basically DOCSIS in 2MHz channel rather than normal 6MHz or 8MHz). DOCSIS does however have the edge in efficiency on higher LOS frequencies especially where multipath (WiMax is OFDM so copes better with multipath) and in wider spectrum allocations as Wireless DOCSIS can use 80MHz FDD or more as multiple 8Mhz channels and even bond them using DOCSIS 3.0.


[Edit Update]

Clearwire to Begin Testing Co-Existence of WiMax and LTE in the Autumn<

Clearwire has come forward in the past to say that they’ve never counted Long Term Evolution (LTE) out, and now with this new report, it looks like they’re strengthening that statement. With tests planned in the Fall of this year, Clear will take a hard look at LTE.

The tests are designed primarily to see how LTE and WiMax live together, within the same market and utilizing the same spectrum. Clear stated that they are going to use hardware from Chinese manufacturer Huawei, utilizing FDD LTE and TDD LTE equipment. And the tests will take place within the 2.4 and 2.6GHz bands, where Clear currently holds assets.

via Slashgear and PhoneScoop<

(Also reported on Engadget and Gizmodo)

But the important thing is to see how LTE can coexist with WiMAX,
ideally from the same base station. Clearwire’s mobile WiMAX base
stations, supplied by Samsung, are capable of doing LTE - but that’s an
untested capability, which is rather the point of this news.


[Edit: UPDATE]<

WiMAX 2 standard, and its theoretical 1Gbps downloads, to be finalized soon <

WiMAX 2 is 802.16m. That standard will
boast theoretical speeds of 1Gbps for downloads, though users are rather
more likely to see something in the 100Mbps range according to Engadget.

But since existing WiMax is close to  the Shannon Limit the extra speed can only come from extra spectrum (wider channels or Channel Bonding like DOCSIS 3.) Such large amounts of spectrum are only in the higher parts of Microwave Spectrum, Strictly LOS (Line Of Sight) so likely the main use will be Point to Multipoint aka "FWA" (Fixed Wireless Access) for Wireless Broadband using outdoor fixed radio set/Aerial. Not Mobile.

It's likely that LTE or existing Mobile WiMax is at the limit of what's realistic for Mobile users, given the reality of spectrum characteristics and 20MHz as a realistic largest channel size. That does give a peak 100Mbps (perfect signal and one user only) dropping to an average throughput of 4Mbps to 10Mbps across the full geographic range of an economically used cell. Unless Mobile Bases are cheap enough to put on every 3rd lamp post, you are not going to have 100Mbps per user!

Update, Confirmed

Intel has announced< a $1.4bn cash deal to buy Infineon’s Wireless Solutions (WLS) business, maintaining the 3G, WiFi, WiMAX and LTE company as a separate venture but also dipping liberally into their expertise for upcoming Intel products and strategy.

via Slashgear<