Farpoint Group –
It's been just about a year since I wrote about ultrawideband (UWB), and there have been a number of interesting developments in this space over that past year. So, I thought this might be a good time to bring you up to date, and to focus for that purpose on applications rather than technology.
But first, a quick review. UWB uses very large amounts of radio spectrum (hence the name) - typically up around 500 MHz. Compare this to the size of a CDMA channel (1.25 MHz), or a UMTS channel (5 MHz), or a Wi-Fi channel (20 MHz), and you can see that this is a large amount of territory indeed. As you know, throughput is very often proportional to the amount of bandwidth available (although there are lots of other variables like transmit power, type of antennas used, antenna orientation, interference, etc.), so 500+ MHz. should allow us to send vast amounts of data. And, indeed, wireless USB or FireWire (IEEE 1304) at 480 Mbps is no problem, and 1 Gbps is also more than feasible.
So, what's the catch? Ultrawideband is limited to power output of about one ten-millionth of a watt. And with that small amount of power spread across a large amount of spectrum, the key tradeoff is in range - as in, not much. UWB thus becomes a wireless room-area or personal-area network (WPAN) at best. We should note here that only the United States has approved UWB at present, and with a few additional restrictions on use. The primary concern is whether UWB causes interference to other signals in the spectrum in which it is operating. I think we have enough experience at present to conclude that it doesn't, and that the other signals are more likely to interfere with UWB (although they almost never will, given the huge amount of bandwidth used by UWB). But no matter - the high throughput we noted above is well-suited to many short-range applications, a few of which I'll cover below.
This column was motivated by a recent invitation I received to speak at the meeting of the International Telecommunications Union (ITU) held in San Diego on "Compatibility Between Ultrawideband Devices (UWB) and Radiocommunications Services." The topic of my talk was "Applications for Ultrawideband Communications." I am, for the record, a big believer in UWB, and I think it is destined to become the short-range radio of choice, assuming a couple of future developments (which I'll discuss below). You can download a copy of my slides if you'd like to get the details. Briefly, though, here are a few thoughts that I shared with the audience:
- Wireless USB (and maybe Firewire or even Serial ATA) are going to be huge. USB has replaced the old serial and parallel connections almost entirely, and there are lots of other USB peripherals. The only real drawback I can think of is that both USB and FireWire cables carry power, and that won't be the case, obviously, using wireless. That's OK; we've got batteries. Wireless USB keyboards and mice will be first, but wireless memory keys are also a distinct possibility.
- Despite the fact that the current FCC regulations prohibit this application, I think toys and games are likely to be the biggest UWB application, assuming the price of UWB chipsets falls below $5 (and I think it will). Imagine networked handheld games (the amazingly popular Texas Hold 'Em seems perfect for this), or toys like a networked Furby, great for those incremental sales opportunities in terms of both more units and software. I think the FCC will change the rules here within a couple of years; the opportunity is just too great. The concern, of course, is over interference; I think it will become clear that UWB causes no interference in the bands to which it is currently directed (3.1 to 10.6 GHz).
- Consumer electronics, primarily in the form of interconnecting home-entertainment components, could also be huge. If you've ever installed a home theater, you know what a hassle it is indeed to connect all of the boxes together. Imagine doing this wirelessly - nothing could be easier. One application frequently mentioned, however, I'm not so fond of, and that's connecting a display to a set-top box. Both are stationary, and both require wires for power, so a cabled solution makes a lot of sense here. But I'll concede that wireless will see at least some application in this case regardless.
- It's pretty clear that UWB is going to replace the current Bluetooth radios. It's a lot faster, and well-suited to the same set of limited-range applications. So, expect all that Bluetooth code to be ported to UWB, and perhaps we'll even see UWB as the third radio in handsets and other wireless computing and communicating devices in the future.
- This brings up, though, the question of possible competition between WLANs and UWB-based WPANs. As is almost always the case, WPANs are cable-replacement technologies, but their basis in radio makes LAN-like structures possible over relatively short distances. It was this vision that helped Bluetooth get into so much trouble, as that technology lacks the throughput to really deliver a LAN-like experience. UWB has no such restrictions with respect to throughput, although it has a similar issue with range. One could, of course, use repeaters and/or meshes and cover a very large area with a UWB-based network. I think the edge here is going to go to the upcoming 802.11n-based WLANs, since they should be able to offer at least 200 to 300 Mbps over time. But I'm also pretty sure that LANs based on UWB will also appear.
- UWB should also be great for location and tracking applications, interconnecting high-data-rate mobile devices like displays (even the head-mounted variety), exchanging data files and pictures in PDAs and digital cameras, and on and on. UWB is very powerful indeed.
And the question, then, is when the international community will set regulations and adopt UWB. That was the purpose of the ITU meeting. We're still a ways away from global deployments, and the battle over standards I noted a year ago rages on regardless. I think, though, it's time for the market to decide. Hundreds of millions of dollars have been invested in UWB companies and technologies, and it's time to get the results into the hands of users everywhere. I expect you'll see wireless USB products before the end of this year, with more applications to follow. I'm looking forward to putting UWB-based products through their paces, and I'll report back to you when we have some in our lab.