Ten years ago on February 14th, the FCC issued new regulations that authorized the use of unlicensed underlay spectrum from 3.1 to 10.6GHz in the United States. The EU, China, Japan, Korea, the UK, Canada, Russia and many other countries soon followed with some restrictions to the upper portion of the US frequency band.
On the 10th anniversary of UWB it is worth taking a look back at the progress UWB has made and the technologies current market status. UWB was authorized during the booming growth of Wi-Fi in laptops, a growth accelerated by Intel’s highly successful Centrino TV and print marketing campaign. Using the lessons learned by Wi-Fi, UWB moved from FCC authorization to silicon rapidly (by wireless technology standards), with the first true worldwide silicon being shipped in a product by Alereon in Q3 of 2008, only six years after the initial FCC rule making.
On the heels of the explosion of Wi-Fi, UWB came to market with hopes of its own explosion focused on eliminating desktop cables and the distribution of video in the living room. Things don’t always work out the way you plan and UWB was slow to be adopted for these primary applications.
There were a number of reasons for the slow adoption of UWB technology. First, the success of Wi-Fi had already established a viable wireless technology that could provide convenience for consumers. True, it was at a much slower data rate and higher power consumption, and it wasn’t capable of replacing a USB cable, but Wi-Fi was widely available, and importantly, it was cheaper. It had taken many years for Wi-Fi to find the right application (laptop to access point for internet access) to refine the cost of the chips, address security concerns, improve the speed, and reduce the power to a point that the technology was ubiquitous. The problem was that UWB hadn’t had a sufficient market gestation period to find the most compelling applications and move down the cost curve.
UWB did in fact learn from the technical history of Wi-Fi. The first products that implemented the initial specifications (such as the Alereon AL5300) did in fact deliver compelling capabilities – a 480Mbps PHY rate at modest range and delivered throughput, depending on protocol, ranging from 220 Mbps to over 300Mbps using a single antenna. Much higher than even today’s Wi-Fi. Moreover, UWB was designed specifically for reliably transferring streaming data, such as video and audio, and to support TCP/IP traffic and USB traffic as well as native data transfers without colliding in the air. Both of these factors made it a much better performing choice in streaming to monitors and HDTVs. In addition, UWB was designed to avoid interference, making it a much better choice for high density deployments (think office cubicles), a common problem for Wi-Fi. With all these advantages, however, the first products were more expensive than Wi-Fi; even if the first UWB chips were generally 10X faster. This cost issue was the primary impediment for broad based initial adoption. UWB did in fact find its compelling application markets fairly quickly – wireless docking and wireless monitors w/docking. However, Wi-Fi had established the price points which consumers were willing to pay for the convenience of wireless and the initial price points of UWB in 2008 were simply too expensive for high volume adoption.
The final cause for slower adoption was the belief that the marketing muscle of Intel alone could make UWB successful in its first attempt. In the fall of 2008 as the UWB community was doing some soul searching as to how to move the market. The world was hit by the largest recession since the invention of semiconductors in the 1950s. Intel chose to abandon its UWB efforts and with the band leader gone essentially every large semiconductor company but one abandoned their internal UWB efforts, followed quickly by the collapse of all but two of the startup UWB companies. This had the predictable domino effect on market perception of the basic technology and of the capabilities of the spectrum itself.
So where are we now? Since 2008 Alereon has reduced the cost of UWB by 60%, the power consumption by 50% and the size of UWB by 85%. Last year Samsung brought to market three very innovative wireless monitors with integrated docks, CA750 23″ and 27″ Central Station, and the CA650 enterprise wireless monitor. In 2012 at CES Samsung announced even more advanced wireless monitors, the new Series 7 Smart Station, which includes support for a wireless connection from an Apple laptop and the ability to tether Android devices. Other major monitor brand names are sure to follow Samsung’s lead. In addition, Alereon has introduced a new 1Gbps capable UWB chip that has 2X the range of the original UWB chips, and recently demonstrated an Android smartphone wirelessly connecting from a belt clip to a prototype wireless monitor – giving the user support for a keyboard and mouse and essentially making their Smartphone a “pocketable PC”.
Looking back UWB’s market adoption cycle is in fact mimicking Wi-Fi’s. Though the standardization effort was faster and the first silicon efforts were faster, the inevitable need to cost reduce, refine and improve the first instantiation of the technology could not be made appreciably quicker. UWB required a time honored learning cycle, just like Wi-Fi and Bluetooth. Hype can be sped up – but consumer feedback and new chips – not so much.
So what is the future for UWB on this 10th anniversary of the FCC regulatory approval? The future in fact is bright. UWB certainly does not compete with Wi-Fi, it compliments Wi-Fi and can do things Wi-Fi wasn’t designed for. My own UWB based wireless monitor has been on my desk for 8 months. The wireless link works just like my old trusty VGA cable PLUS I have no docking station on my desk; all my USB peripherals are attached to the base of my monitor and are out of sight behind the back of my desk. Wi-Fi simply can’t do that. With the availability of 3 GHz of unlicensed spectrum UWB will continue to improve. 480Mbps shipping today, 1Gbps soon, 2Gbps in 2013 and ultimately much faster yet. 130nm to 90nm to our current 65nm chips and designing our next products for even deeper lithography to double the data rates, lower cost, add USB3.0 support, etc., etc.
Though lithography and consumer desires continue to change, physics doesn’t. UWB continues to benefiting from Wi-Fi’s ubiquity and the resultant interference and congestion that reduce Wi-Fi’s performance. New customers in industrial, medical, automotive and military applications are engaging with UWB technology due to its unique streaming characteristics and immunity from Wi-Fi interference and congestion. Ultimately, UWB will be shipping in 100s of millions of units. The ramp to these volumes won’t have taken any longer than it took Wi-Fi or Bluetooth. However the widely held perception that hype alone can make a semiconductor market happened sooner has been proven wrong once again. Like good wine, good standards, like UWB, get better with age.