Wireless Systems Design

Good News About Ultra Wideband

The good news is that ultra wideband (UWB) wireless is finally here. The bad news is...well, I can’t think of any bad news. UWB chip vendors finally have working silicon. Now they are announcing their FCC approvals and WiMedia certification and registrations, as well as identifying some initial customers. Looks like we will actually be seeing some real UWB wireless products this coming year.

The FCC originally approved UWB in February of 2002. FCC approval resulted in a flurry of activity to develop the technology and get it to market. Startup Xtreme Spectrum beat everyone to market with a direct sequence (DS) pulse type UWB that worked great. Xtreme was quickly bought by Freescale (Motorola Semiconductor at the time) and developed a very workable product.

The usual standards battle ensued next. This was the IEEE 802.15.3a war of the semiconductor giants for a UWB standard. That eventually turned into a standoff between the DS UWB faction (mainly Freescale) and a whole host of others promoting a multi-band orthogonal frequency division multiplexing (MB-OFDM) version. The whole process ended after years of fussing with a disbanding of the IEEE 3a Task Group. No standard emerged and the groups went their own way. Most of the chip companies formed and joined the WiMedia Alliance to define the OFDM version, which has since become a standard of its own. Most of those making UWB chips such as Alereon, Realtek Semiconductor, Staccato Communications, TZero, WiQuest, and Wisair and others use that standard.

The WiMedia standard defines 14 bands 528 MHz wide in the 3.1 to 10.6 GHz range. Only the first three bands (from 3.1 to 4.9 GHz) are used in current chips, but that is sufficient to create an OFDM radio that can deliver from 53 Mbits/s to 480 Mbits/s at a range up to 10 meters.

UWB has always been promoted as the way to transmit video wirelessly over short distances, and that is still a potential application. However, most UWB chip companies are making wireless USB products. They let you plug and play any device with a USB port with a UWB dongle, hub or other interface. Initial UWB products will use USB. PC peripheral connections are the main target, but other uses will include wireless connections to digital cameras and other devices.

While UWB definitely has the speed to transmit even uncompressed high definition video streaming, many doubt that its range is sufficient for consumer use. Its low power nature does not let it punch through walls or go beyond a few meters at it highest data rate. UWB innovator TZero has a MIMO version of UWB that helps extend the range that may be a vision of things to come in the way UWB succeeds in video. The 802.11n Wi-Fi gang swears that their faster version of Wi-Fi will be the winner in the home wireless video space. It has the speed and the range thanks to MIMO. And even then some other standard may emerge to fill that need. A newcomer called Wireless HD will operate in the 60 GHz band and could capture a bit of that business with its potential to support data rates up to 5 Gbits/s or so.

After all the work of the past years, UWB is set to be on the market next year. You can expect to see a variety of products—some of which will promote wireless video. Wireless USB applications will dominate, but expect to see some interesting new applications. In a recent visit to Alereon in Austin, Texas, I saw CEO Eric Broockman show off their USB dongle talking to a prototype version of Kodak's V610 digital camera with embedded UWB. It downloaded huge photo files almost instantaneously at a distance of several yards. Consumers will love that high speed as much as the fact that the two wireless devices find themselves and link up automatically without any consumer knowledge or set up.

Another interesting development is the Bluetooth SIGs recent adoption of UWB as the next step in the Bluetooth PHY roadmap. Bluetooth's 1 Mbit/s max data rate is certainly sufficient for audio applications like cell phone headsets. And their EDR (Enhanced Data Rate) version boosts speed to 3 Mbits/s for stereo and audio streaming. With the selection of the MB-OFDM version of UWB, the next generation of Bluetooth products can support speeds up to the 480 Mbits/s limit. Bluetooth chip leader CSR recently announced their forthcoming UWB product that targets mobile products rather than USB.

Eventually we will see UWB move up into the higher parts of the allocated spectrum. The USB Implementers Forum plans to add support for UWB in the spectrum beyond 6 GHz in the Wireless USB specification in 2007. This should really expand the options for manufacturers in specific geographic regions where the current chips are not allowed. Making chips to work in the 5 to 10.6 GHz band is not that easy, but they will be on the way in another year or so. While that will shorten their range even more, they won't have to worry much about interference to or from other wireless services.

So things are really looking up for UWB. Like all new wireless technologies, it takes time to develop and fine-tune. Looks like we will see UWB find its niche mainly in the US. Pretty soon UWB will find its way into other countries as their regulations permit. It will be interesting to see how it fares in the forthcoming home video networking battles that are just beginning as IPTV rolls out in the coming year. And who knows, we may actually see something like a wireless IEEE 1394 option.

TAGS: Freescale
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.