Intel gets behind 'Wi-Fi on steroids'

Analysis: Will the launch of Intel's new mobile WiMax chip eventually push the broadband wireless technology into the same realm as Wi-Fi? The chipmaker's European development manager Chris Beardsmore certainly hopes so

Intel was the powerhouse that drove Wi-Fi towards ubiquity. Now it wants to do the same with WiMax, the long-range, high bandwidth technology that has been popularly described as "Wi-Fi on steroids".

Fixed WiMax is already starting to reach the market in a behind-the-scenes sort of way, but many have suggested that the technology will only truly take off in its mobile form, which has taken longer to develop.

That momentum around mobile WiMax will receive a much-needed boost from the announcement on Monday of Intel's Rosedale II chip, which the company hopes will soon be as standard in laptops and handheld devices as Wi-Fi is today.

But mobile WiMax faces significant challenges in the UK, particularly from fast-advancing 3G technologies and the mobile operators such as Vodafone and T-Mobile who have already invested heavily in them.

ZDNet UK spoke to Chris Beardsmore, Intel's marketing development manager for the EMEA region, about the road ahead for mobile WiMax.

Q: How big is the demand for mobile WiMax in the UK, especially with the rise of Super 3G technologies?
A: In terms of consumers and users, our sense generally is that the need for bandwidth never decreases, and they want mobile data — Wi-Fi set the benchmark for doing that./>

HSDPA [Super 3G] is a great technology for giving people mobile data, and they are in a position to be able to deploy networks today and offer that. Possibly the issues they have are in terms of bandwidth, quality of service and cost per bit.

WiMax fits into a space where you need more bandwidth than you can get from HSDPA — we sit as a complementary technology there, between Wi-Fi and HSDPA. In terms of those people who are looking at it, we have people like Pipex, who own spectrum and have a data-business case, so it's fairly logical for them to move into that space.

The likes of companies such as BT, who have fixed assets but very limited mobility, are also looking to expand into the nomadic data market with some of the things they're doing, such as wireless cities [an initiative to create metropolitan "hot-zones" across the UK].

They are obviously very interested in WiMax as another piece of their plans. Whether Rosedale II intersects with their plans, I'm not certain at this point. Rosedale II might fit in with their DSL fill-in, for example where there's no good copper infrastructure — they've looked at WiMax as a DSL fill-in tech for the 2-3 percent of the population with no access to DSL broadband.

Did Rosedale I, Intel's fixed WiMax chip, live up to expectations in the market?
I would say so. We were very pleased with the uptake of Rosedale I. It allowed vendors to bring the costs of client devices down, and we're very pleased with that.

What we're seeing with Rosedale II is, not only have the vendors who bought Rosedale I said they'll buy it, we've also seen names such as Alcatel enter the market.

Tell us more about Rosedale II…
Rosedale II is our first silicon that supports 16e [mobile WiMax standard 802.16e). It's also backward-compatible with 16d [fixed WiMax standard 802.16d], which gives us a migration story from 16d networks being deployed today

We took the existing Rosedale I design and looked at what we would need to add to it to make it support 16e. We made some internal hardware changes...

... in the silicon, but we found we could actually make it pin-compatible, which was great [it solders down onto same piece of fibreglass]. We also found that we could keep the 16d compatibility, making it a bridge device.

We introduced a radio component called Ofer-R [dual-mode WiMax/Wi-Fi technology], and that's a single-chip radio device that goes with Rosedale and drives the costs and integration down for modem devices.


How would you respond to Qualcomm's Europe president, Andrew Gilbert, who recently said he found it hard to see where mobile WiMax would fit into the market by the time it became available?
It's good that HSxPA techs are progressing, but if you look at where they are in terms of bandwidth and cost per bit, we believe WiMax has a significant advantage in both those areas, and also in terms of quality of service and reliability.

Qualcomm obviously has intellectual property rights in the HSxPA and CDMA technologies, and one of the advantages with WiMax is that the intellectual property rights are spread fairly evenly across the technology, so no one company has a controlling interest in royalties, so that reduces the costs significantly.

Qualcomm has also made an investment in OFDM technologies [which form a basis for mobile WiMax], and if you look at most of the people looking at "4G", they all agree it's based on OFDM. We think WiMax is there in terms of OFDM and is the basis for a solid data network — it's possibly closer than you think.

We've definitely seen technology and prototype demonstrations of mobile WiMax. We're seeing trial deployment of mobile WiMax in South Korea. Field trials and early commercial trials [will reach Europe] towards the end of this year.

The GSM Association is opposing technology neutrality in key bands of spectrum — how is the fight for usable mobile WiMax spectrum going?
That's possibly a strategy they're using, but it's not true of all the operators, by the way.
What Intel is doing is lobbying for technology neutrality and access to the 2.5GHz spectrum across the EU and worldwide. Regulators are becoming a lot more open to that discussion.

Obviously they move relatively slowly, but we're seeing acceptance within the regulators for the possibility of technology neutrality in a 2.5 band, and a compromise position of using the centre gap frequencies [earmarked by the GSM Association for 3G extensions] is possible.

Do you think there is a valid case for reserving those frequencies for 3G services?
3G was mainly designated for data traffic, but is largely being used for voice traffic with a small portion of data on top of that. There is an argument saying they should be allowed to provide services in that spectrum [but] whether they should be allowed to preclude other people from using that spectrum is a different matter.

Let the market decide which is the best technology for providing those services. They do have a reasonable amount of spectrum today and I don't believe that spectrum is being used to anything like its capacity at the moment, so there's certainly growth opportunities there.

WiMax is very spectrally efficient — so maybe when we look at long-term evolution [future incarnations of 3G technologies] and 4G...

...that may be a discussion point, but again the technology-neutrality point of view says that if technology exists, and it's a good fit for providing services, then it should be permitted to do that./>

Long-term evolution and 4G are all looking at OFDM as a technology, which forms the basis for WiMax already — what the intersection part of those two viewpoints are, I don't know.

How great has vendor interest been in Rosedale II?
We're sampling customers now. We have 10 who've announced publicly — they've all committed to build products based on Rosedale II. We're expecting a production product later this year.

And will that product be WiMax Forum certified?
It's a question of when companies put our products through the certification process based on our silicon.
It doesn't stall the deployment at all — it’s great to have industry and ecosystem support behind it.

What's the way forward for mobile WiMax now that you've announced Rosedale II?
Rosedale II fits into the residential gateway market. We will next see PC cards for laptops, and then integration into the Centrino platform… probably at the back end of 2007, or the beginning of 2008, however we need to intersect to the point where there are actually networks to run those devices on.

Our main focus is laptop technology. Some of the other silicon vendors, such as Samsung, are looking at trialling handheld devices — some trials are happening already in Korea. Consumer electronics will probably come slightly later than laptops due to their lower power and smaller form factor.

How is the relationship between mobile WiMax and its Korean cousin, WiBro, working out?
We have convergence on WiBro. They added some great pieces to the WiMax specification and we helped them with some of the work they were doing. WiBro is one of the profiles of the WiMax Forum — it was a great win-win situation.