See also: Electronics Weekly's Focus on Wireless, a roundup of content related to wireless communications.
Over the last month or so, the discussion around Ultra Wideband (UWB) has been about recent developments in the companies developing the technology. WiQuest, an early leader in the UWB space, announced at the end of October that it was unable to raise a new round of funding and had to shut its doors.
On the heels of that news were reports that Intel had decided to shutter its Ultra Wideband Networking Operations - it had actually decided to do so months prior to the fanfare.
There have been claims that “UWB is dead” or at least has one foot in the grave. Others have touted that market attrition and consolidation is not only expected, but required, and the promise and value of UWB as a high-speed, low-power, short-range wireless technology remains unchanged.
Let’s start at the beginning
In February 2002, the US FCC authorised the unlicensed use of UWB in the 3.1GHz to 10.6GHz spectrum.
A battle ensued in the IEEE standards committee between two camps to adopt their version of UWB.
The Multi-Band OFDM Alliance - since merged with the WiMedia Alliance - had defined and promoted a multi-band OFDM technology, while a group known as the UWB Forum promoted a direct sequence pulse-based radio design.
Without going into details of the differences of the technologies or the politics involved in the standardisation effort, ultimately neither was adopted due to both camps’ inability to acquire the necessary votes.
In the spring of 2006, the IEEE task group was officially disbanded.
The UWB Forum is now defunct, while the WiMedia Alliance continues to grow with over 350 member companies across multiple market segments.
WiMedia UWB technology specifications have been adopted by the USB-IF for Certified Wireless USB as well as adoption as official ECMA standards.
The Bluetooth SIG has also announced collaboration with WiMedia to develop next generation high-speed Bluetooth technology.
So a de-facto standard for high-speed wireless personal area networking (Wireless PAN) has emerged in the form of WiMedia UWB.
And as history has shown us with technologies such as wired USB, which is arguably the single-most successful connectivity technology ever introduced, the lack of an IEEE standard is not a barometer for failure or success.
Why UWB?
Driven by innovations in mobility, power, and storage capacity, users continue to have a voracious appetite for the latest and greatest in consumer electronic gadgets.
Add to that a new world of high definition content and social networking, and it’s no wonder that there is no end in sight to the growth of digital electronics, and Wireless PANs will play a key role in how people access, share and enjoy this content rich environment.
A Wireless PAN has a typical range of 10m, and enables simple, secure connections to allow digital devices to communicate with one another at data rates up to 480Mbit/s.
These connections may be mobile - where a friend shares his digital photos with another friend via their handsets.
They may be fixed - in a home office where the printer, speakers, monitor and other peripherals maintain a wireless connection with the PC.
Whatever the usage scenario, there is a need for a high-performance, low-power wireless solution that can also achieve the price points to enable high volume applications such as PCs and mobile handsets.
These are requirements that are not well serviced by existing wireless technologies.
As a comparison: UWB can transmit data up to 200 times faster than existing Bluetooth technology, and is 5-10 times more power efficient than Wi-Fi.
Also, UWB also does not operate in the crowded 2.4GHz and 5GHz bands, reducing the chance of interference.
Why isn’t the world flooded with UWB-enabled devices?
Even though the FCC had opened up spectrum for UWB operation, the rest of the world still had restrictions. This prohibited device manufacturers from offering a worldwide compatible solution, a requirement for most of the major market segments.
The good news is that today, in addition to the US, UWB operation is allowed in Europe, Japan and Korea with expectation that China will grant approval by the end of the year.
The early “Gen 1” solutions delivered by IC vendors, which are ultimately the building blocks for the end products, only supported operation in the lower-end of the spectrum (known as Band Group 1) so did not support a global footprint. However, we are now seeing solutions on the market that support multiple band groups (specifically groups 1, 3 and 6) enabling worldwide-compatible products.
Another issue with the early Gen 1 solutions, as is typical with any nascent technology, is that the size, power and especially the price points were just too high to enable mass adoption in volume markets.
On top of this, due to backward compatible wire adapter protocols for wireless USB, early product performance was less than impressive, delivering around 25% on the true capability of the technology.
Though wire adapters allowed for simple addition of wireless USB capability with little or no changes to existing hardware, customers and users saw the performance as a restriction of the technology in general.
Today we are seeing native implementations of wireless USB products that can achieve greater than 200Mbit/s, and that Gen 2 solutions are hitting the streets that meet the cost, size and power requirements of PC, peripheral, consumer electronic and mobile handset markets.
One such solution is Staccato’s Ripcord2 65nm CMOS single-chip solution which has been certified by WiMedia in Band Groups 1, 3 and 6.
Markets in the past (we’ll look at Bluetooth and Wi-Fi in a second) have required similar game-changing technology to move from the early adoption phase to the high volume phase, and eventually to the ubiquity phase.
Long history, short memories
It is not surprising to see attrition in this or any other market these days, and I believe that the current downturn has had a hand in accelerating the attrition of UWB companies.
There is a long history of technologies materialising in a very similar manner.
If you look at any relevant technology, say Bluetooth or Wi-Fi for example, you will find common stories of failed start-ups, stalled technology, and interoperability issues.
Every technology goes through these growing pains, and a pattern can be recognised where the winner, although not first to market overall, was first to provide a solution that meets the requirements of the market.
In the Bluetooth space, among the two or three dozen start-ups, CSR emerged as the leader. For Wi-Fi, where there were even more players, Atheros was the winner.
UWB is going through similar growing pains, and existing vendors now have solutions that deliver on early promise.
The claims made in certain articles about the imminent demise of UWB are not correct and not unique.
In 2003 an article claimed “Bluetooth Is Dead”. Last I checked, Bluetooth shipments now tops 1 billion units per year.
In the case of Intel UWB specifically, it’s important to note that the Ultrawideband Networking Operations was a start-up of sorts that was funded by Intel’s New Business Initiatives arm, not by a product group.
When a business review was recently held, Intel decided that UWB is not considered a piece of their core business, and would prefer to buy the technology if they need it in the future.
Few remember that Intel also started USB and Bluetooth businesses in similar fashion. The USB business was eventually sold off to Cypress and the Bluetooth group closed down.
This is quite typical of Intel for non-core businesses. They drive the technology during the incubation stages, but then allow outside vendors to take it over after it matures.
So if Intel is the barometer for the success of complementary wireless technologies, I like UWB’s chances.
And Intel continues to keep a close eye on these technologies through involvement in the industry consortiums such as the USB-IF, as well as through investments through their Intel Capital arm.
In the UWB space, Intel has also invested in Staccato and Wisair, having participated in a Staccato funding round last month.
The fact of the matter is, WiQuest and Intel were unable to execute on their next generation solutions to meet the requirements of the market, so their investors pulled the plug.
The silver lining in the shake-out is that there is more business to be had among the remaining vendors.
The Road Ahead
The future of UWB is promising. There simply is no better technology to transfer data wirelessly at high-speeds and low-power. And with the introduction of Gen2 solutions, we are likely to see compelling new usage models in a broad range of products, from PCs and peripherals to consumer electronics and mobile handsets. The WiMedia Alliance is actively pursuing a roadmap that will extend the data rates beyond 480Mbps, allowing the technology to keep up with the trends and needs of the market for a long time to come.
“The reports of my death have been greatly exaggerated.” Mark Twain
Jeff Chang is v-p of marketing at Staccato Communications
See Broadcom's views of the development of a market for UWB.