Ultra-Wideband,
the Future of
Of all the currently emerging short-range wireless solutions, which one is poised to take us into the next era of computing? This paper intends to show that the answer is Ultra-wideband or UWB. UWB is not perfect; the FCC has not yet approved UWB for commercial release because of possible interfere with other devices. Still, no other wireless solution has the promise of Ultra-wideband.
In a perfect world, without wires we could, “send a lot of data, very far, very fast, for many separate uses, [and] all at once.”[1] Unfortunately, we do not live in a perfect world; there are physical barriers that will not allow all of these goals to occur simultaneously. It is necessary to minimize one or two of them to maximize the others. In the past, the main requirement of wireless communications was to travel very long distances (like transatlantic radio) because there was an absence of a suitable infrastructure for crossing these expanses. We achieved those exceptional distances at a cost of everything else. Today, we have a wired infrastructure that reaches almost every part of the globe; therefore our wireless mediums need only travel far enough to tap into the current infrastructure. Our current cellular systems are an excellent example of this idea. The only distance covered, without wires, is from a cell tower to a phone and back, everything else, is wired.
With the current proliferation of high-speed wired Internet access, short-range wireless (SRW) is poised to step in, as cellular did with the phone system. By definition, short-range wireless passes on range in favor of the other attributes; SRW systems have a range of 100 meters or less.[1] The current leading SRW technologies are Bluetooth, IEEE 802.11b, IEEE 802.11a, and our focus, Ultra-wideband.
Uses of Short-range Wireless
Short-range wireless has a wide variety of uses. Depending on which technology is implemented, SRW can also be used to create a personal area network (PAN), with a range of zero to ten meters (Bluetooth), or a local area network (LAN) with a range of zero to 100 meters (802.11b). Bluetooth’s current main use is to create a PAN between devices like cell phones, PDA and PC’s. Combining many small range systems in one building, such as a mall, airport, or hotel can provide either Internet access or access to a proprietary system (for example, to check into your hotel room without waiting in line) to anyone that enters the building.[1]
If too many systems are operating in the same area at the same time (as in the previous example) they will degrade or even nullify each other’s capacity. This leads to a new measurement beyond simply data rate or range. It is necessary to look at the number of systems that can safely be used in the same area, at the same time. Then we need to look at the “spatial capacity”[2] of the technology. Spatial capacity can be determined by calculating total data rate supplied by all the systems in a given area and then dividing by the area to arrive at bits per second per square meter.
802.11a
operates in 5GHz U-NII band; at this frequency level there is about 200 MHz of
useable spectrum. It has a projected
operating range of 50 meters. 802.11a
has been measured to have a peak speed of 54 Mbps. As many as twelve 80211.a systems can operate
at the same time within a given 50 meter circle. Twelve operating systems deliver a total data
rate of 648 Mbps. 648 Mbps spread over a
50 meter circle delivers a spatial capacity of 83 Kbps.[2]
802.11b
operates in the 2.4Ghz ISM band. This standard has a larger operating range
than 802.11a at 100 meters, which is better suited to form a LAN than a
PAN. Although 802.11b has the larger
operating range, there is less usable spectrum available to it at only 80 MHz. This available spectrum allows for three
systems to interoperate in the same 100 meter area. The top speed of 802.11b is 11Mbps. The three simultaneously operating systems
then have a total speed of 33 Mbps. 33
Mbps calculated over the 100 meter coverage produces a spatial capacity of only
1 Kbps.[2]
Bluetooth
also utilizes the 2.45 GHz ISM band.
Unlike the other technologies, Bluetooth is designed to be ad hoc, which
means that a connection can be established between any two Bluetooth devices; most
systems have a base through which all devices must communicate.[3]
Since it
does not utilize a base site for communications, Bluetooth forms ‘piconets’. Although
there is no defined base, for many items to communicate something has to
coordinate them. A piconets
is formed when at least two devices attempt to communicate, one is named the
“master” and the other “slave” such that the master defines frequencies and the
timing of signals between all slaves within its’ piconet.[3] Ten such piconets can safely operate at the same time within a 10
meter circle; each has a data transfer rate of 1 Mbps. That allows for a total maximum of 10 Mbps
inside the given 10 meter circle, which translates to a 30 Kbps spatial capacity.[2]
UWB is not limited to a specific frequency although it generally falls between 1.5 GHz and 4.0 GHz.[1] This covers a wider range of frequencies than does any of the other technologies as will be discussed shortly. Ultra-wideband is defined as “any radio system that has a bandwith greater than 25 percent of its center frequency, or greater than 1.5 GHz.”[1] This is a much wider channel bandwith than is supported by any of these other standards. This is one of the great advantages of UWB, as we will discuss later.
Ultra-wideband systems have been tested and have achieved data rates of 50 Mbps at distances of 10 meters. Projections indicate that 6 of the tested systems could operate simultaneously within a 10 meter circle. This gives us a maximum of 300 Mbps within our circle, which coincides with a spatial density of 1,000 Kbps. Therefore, UWB has a maximum capacity more than 12 times its closest competitor, 802.11a, which is only 83 Kbps.
As mentioned briefly before
UWB utilizes a much wider channel bandwith than any
of the other mentioned technologies.
This is what allows UWB to attain such a high spatial capacity. A look at the Hartley-Shannon law
demonstrates how this is possible, “C = B log2 (1+ S/N)”. In this equation C equals the maximum channel
capacity, B is the channel bandwith in hertz, S is
signal power and N is noise power, both in watts. We can see that as the bandwith
of the channel grows the capacity also grows, linearly. Since the other technologies are much more
limited in their available channel bandwith, their
capacity cannot match that of Ultra-wideband.[2]
The nature of the UWB signal allows it to, theoretically, be interoperable with almost any narrow-band technology. This is necessary because the large range of operating frequencies that are utilized by UWB overlaps with many already assigned frequencies, including those of other SRW technologies. UWB operations in the vicinity of a “narrowband” receiver will simply act to raise the floor noise of the system with which UWB overlaps.[2] It is a combination of both the large signal bandwith and the low power level of a UWB transmitter that create this effect.
The power emitted from a UWB transmitter is so low that it falls within the allowable limits for “incidental radiation” as defined by FCC Part 15.[1] Incidental radiators are electronic devices that unintentionally emit signals that could possibly interfere with other devices; incidental radiators cannot radiate power higher than a specific limit or they must be regulated separately from FCC Part 15. Most electronic devices that are not intentional radiators, like transmitters, do still radiate some power and hence are incidental radiators. UWB systems can operate with 50-70 milliwatts of power; this is one ten-thousandth of that which is used by the average cell phone.[4] Increasing the power output would greatly increase the distance and the data rate for Ultra-wideband but then UWB would no longer be viable given that the frequencies needed by UWB are already assigned.
Although it limits our data rate and range, the low power has many advantages. Obviously, as was just demonstrated, the low power helps to efficiently use the frequency spectrum by giving UWB the ability to interoperate across many currently used frequencies. The low power requirements can greatly increase battery life for devices that communicate using UWB over other SRW technologies. Power consumption can also play a financial role when SRW is used on a large-scale basis; the less power the computer systems need to transmit data, the less power the company has to buy. A UWB system that was tested with higher than normal power output, allowing data in excess of 100 Mbps, still only used 200 microwatts of radiated power. This is about one-fifth the amount of a low-power Bluetooth link.[1]
UWB transmitters and receivers are also simpler than traditional radio transmitters as used in the other given SRW technologies. This will allow for easier mass production and maintenance. The UWB system does not rely on the standard super-heterodyne circuit or a variant thereof, used to create a radio signal, as does Bluetooth. Architectures based on these circuits can be very complex because they require crystal oscillators and the addition of power amplifiers, and synthesizers.[2] Ultra-wideband systems merely need a single chip capable constructing and modulating the pulses through which UWB data is sent.[2] When the rate at which chips are improving, and at the same time their price is dropping, is taken into account, choosing a single chip over a complex architecture seems like an easy decision.
The biggest
obstacle to a commercial release of UWB technology is FCC regulation. Even though ultra-wideband technology
generally falls within the limits for incidental radiators there are two main
objections. Although UWB does fall under
the acceptable noise levels for incidental radiators, UWB is not an
incidental radiator; it acts as an intentional radiator. The FCC is currently considering specific
rule changes that would allow UWB transmitters within the scope of Part 15 even
though it is an intentional radiator.
The FCC has much to consider because there have been many filings in
response to the Notice of Inquiry (docket 98-153) filed in 1998.[5] The FCC is also considering special
regulations to lower the limits for incidental radiators in certain frequencies
to protect the critical GPS system and other devices that operate below 2 GHz
and this is part of the range covered by UWB.[1,2] Although tests have shown that, when
properly implemented, UWB can fall under the limits for incidental radiators,
it does cause some interference.
Testers have concluded that this level is acceptable, but that can also
depend on how one defines acceptable in each situation.[4]
With the incredible spatial capacity gap between Ultra-wideband technology and the rest of the SRW world, it appears that UWB is the future of wireless. The technology has been progressing even though FCC approval is still pending; the peak speed of UWB only 2 years ago was 2.5 Mbps. FCC Chairman Michael Powell informed Congress that the FCC would address this issue before the end of 2001.[4] If and when FCC approval has been granted expect to see UWB flood the market quickly, and just as quickly it will likely dominate the wireless market.
References
1. Leeper, David G. “A Long Term
View of
2. J. Forester et al., “Ultra-Wideband Technology for Short or Medium Range Wireless Communications,” Intel Technology J., Q2, 2001, http://intel.com/technology/itj/q22001/articles/art_4.htm (current September 2001)
3. Haartsen, Jaap C. “The Bluetooth Radio System,” IEEE Personal Communications, IEE Computer Society, February 2000.
4. Cox, John. “Ultrafast wireless
technology set to lift off,”
5. M. Rofheart, “XtremeSpectrum Multimedia WPAN PHY,” IEEE 802.15.3 Working Group Submission, La Jolla, Calif., July 2000, http://grouper.ieee.org/groups/802/15/pub/download.html (current September 2001)
Home Outdoors
Various School Projects Rat Babies Wacky
Test Results Contact Kevin McMahon