By Steve Stroh
Editor, FOCUS On Broadband Wireless Internet Access
http://www.focusonbwia.com
This article is excerpted from the July/August, 2001 issue of Focus On Broadband Wireless Internet Access and is offered as an example article. In December 2004, this HTML version was cleaned up, including very minor formatting changes (but no changes to the text) and reposted to be accessible from Slashdot.org and other links at its original URL - http://www.strohpub.com/0701feat.htm.
The phrase ELE / Extinction Level Event, entered the popular consciousness several years ago as a result of the popular movie Deep Impact. In the movie, an enormous asteroid is observed to be on a collision course with Earth. The asteroid is sufficiently large that an impact on Earth will cause catastrophic effects, mostly a dust cloud that will block sunlight for many months if not years, triggering the death of plant life, and soon after most animal life.
The term ELE came to mind as I read about a new lighting technology from Fusion Lighting, Inc. (www.fusionlighting.com) that uses microwave energy in a new, very high-efficiency lighting system, dubbed "RF [Radio Frequency] Lighting".
An August 6, 2001 article in the Wall Street Journal titled "Energy-Saving Light-Bulb Maker Battles With Satellite-Radio Firms For Bandwidth" describes a battle-in-the-making between Fusion Lighting, Inc. and two companies that plan to offer satellite-based broadcast radio - Sirius Satellite Radio, Inc. and XM Satellite Radio. At issue is the amount of interference that Fusion's new devices would cause to the satellite radio broadcasts at 2.320-2.345 GHz, which are considerably removed from the spectrum where Fusion's devices operate - 2.400-2.4835 GHz. The satellite radio broadcasters have concluded that Fusion's devices, as proposed, will cause substantial interference to their transmissions.
Left unmentioned in the WSJ article, and only now beginning to be noted by many users of the 2.4 GHz band is that if the Fusion devices are capable of causing such trouble for satellite radio broadcasting... what would the effect be to communications users of the 2.4 GHz band, where the Fusion devices will be operating?
2.400-2.485 GHz in the US is used by two very different types of equipment. The older, more well established use of the band is for Industrial, Scientific, and Medical equipment (ISM) such as household and industrial microwave ovens. Operations of such devices are governed under the FCC's Part 18 rules. Basically, Part 18 devices are expected to radiate only - not receive and thus, are not communications devices.
The second major use of the 2.4 GHz band is for license-exempt communications equipment governed under the FCC's Part 15 (15.247) rules. Because the Part 15 rules specify "robust" modulation techniques such as Frequency Hopping Spread Spectrum (FHSS), Part 15 and Part 18 devices can generally co-exist, for example microwave ovens in a household rarely operate for more than a few minutes at a time, so cordless phones and wireless networks operating in the 2.4 GHz band can continue to operate. Conflicts were anticipated when the Part 15 operation was first envisioned, and the following requirement was levied on Part 15 devices:
(1) [Each Part 15] device may not cause harmful interference, and
(2) [Each Part 15] device must accept any interference received, including interference that may cause undesirable operation.
Basically, buyer beware / use at your own risk.
Fusion's RF lighting system is a good example of a Part 18 device. RF technology is useful for any number of purposes other than communications, and the FCC and industry recognized this and set aside various chunks of spectrum for industrial use. The problem comes that the 2.4 GHz band is now used by an incredible number of number of communications devices - cordless telephones, wireless Internet access networks, wireless Local Area Networks, and soon all manner of simple devices with Bluetooth embedded in them- with an accumulated investment of (at minimum) billions of dollars.
How we got to this situation is that there was steadily increasing demand by various companies for spectrum for wireless networking and related applications, but there were no large swaths of spectrum that were suitable. There were demonstrable needs for wireless LANs in warehouses and hospitals. There were demonstrable needs for temporary wireless links. Eventually the FCC decided to offer a Faustian bargain: Industry could use the existing ISM spectrum if they adhered to certain technical limitations with no expectation of protection. It was clearly in the FCC's mind that there would relatively few Part 15 devices, and that for the most part they had heard the last from Industry. The Part 15 rules were tough, and it would be extremely challenging to make radios work under such conditions and they'd have to develop newfangled spread spectrum technologies that were previously used only by the military and developed at great cost.
But, Industry found the Part 15 "deal" to be perfectly acceptable. Industry understood the Part 15 deal better than the FCC did. Industry's major goal was to be able to offer wireless devices that did not require a license from the FCC, so that such wireless devices could be sold over the counter - to anyone, everywhere. Industry foresaw that there was a market for millions of such devices (I doubt that Industry, at that point, projected that such devices would rapidly number in the billions.)
That there were technical obstacles to overcome... well, that was just a barrier to entry for potential competitors. As we've come to expect, where there is a demonstrated demand, technology can overcome, and that's exactly what happened. The biggest factor that made the difference is the rapid increase in capability of application-specific integrated circuits, and digital signal processors. Taken together, spread spectrum radios could be built, at affordable prices, that met the FCC's Part 15 rules. Gradually, an entire Part 15 industry evolved, far beyond the wildest imaginings of the FCC.
What Will Happen? There are a number of factors at play, and very high stakes, so there are a number of possible scenarios. The first scenario is that, quite apart from its effects within the 2.4 GHz band, the effects of the new Fusion Lighting devices outside of the 2.4 GHz band must adhere to existing regulations. It's difficult to ascertain from what has been published to date (particularly when Fusion Lighting is being very circumspect with potentially damaging details of its proposed product), but it appears that Fusion Lighting claims to meet the "out of band emissions limits" for Part 18 devices.
The counter-argument from the satellite radio broadcasting companies is that even if Fusion Lighting's proposed products are within out-of-band emissions limits, their transmissions are still being impacted. To which Fusion might be expected to reply (to the effect of) "If a satellite radio broadcasting system is too precarious to deal with other signals that should have been expected, then you didn't do your homework."
The satellite radio broadcast companies' position is that this kind of interference has never been previously been an issue. Etc. You can understand why this is such a hot issue at the FCC.
But, within the 2.4 GHz band, if a Fusion Lighting device is activated, it will severely impact the use of all manner of Part 15 devices in the area around it. For example, 802.11b is becoming very popular in both large and small companies, and becoming even more popular for home use (because to hook the kid's computer up to the cable modem doesn't require any new wires). Cordless phones are also increasingly using 2.4 GHz. What happens when a nearby gas station installs RF lighting and all 802.11b devices and 2.4 GHz cordless phones for a mile in diameter stop working?
The RF Lighting issue is quite the dilemma for the FCC, which was hoping that Fusion Lighting would be willing and able to modify their device so that it wouldn't cause interference to communications equipment. But that appears unlikely, and Fusion Lighting appears to be within its "rights" to apply for an FCC Part 18 certification to begin manufacturing. But if Part 18 certification is granted and RF Lighting devices become widespread (and it appears very likely that they will, given their inherent energy efficiencies), is the FCC willing to "sacrifice" much of the utility of the 2.4 GHz band in exchange for one company's (at the moment) product?
The group likely to be most severely impacted by Fusion Lighting devices are Internet Service Providers that are using wireless equipment to connect to their customers. The vast majority of Wireless ISPs (WISPs) use equipment that operates in the 2.4 GHz band. Some equipment is purpose-built for ISP use, and many others use modified Wireless Local Area Network (Wireless LAN) equipment. What all WISPs have in common is that their signals are relatively "fragile". The FCC's Part 15 rules apply equally to equipment used by Wireless ISPs, so the ISPs compensate for low transmitted power with high-gain, directional antennas. This approach allows them to have enough "signal margin" to achieve a reliable link, but if a source of interference appears nearby, the link will likely be disrupted.
It may well be possible to overcome interference in the 2.4 GHz band from RF lighting devices, but doing so won't be inexpensive or easy. For example, link margins can be improved by building multiple hub sites with short paths instead of just a few hub sites with relatively long paths. Another approach is to buy better 2.4 GHz band equipment that is more robust, such as that offered by WIMAN Systems (www.wiman.net).
A long term solution to interference issues in the 2.4 GHz band is to begin using equipment that operates in the 5 GHz band. In the US, there is a total of 300 MHz of spectrum available for license-exempt wireless devices at 5 GHz. 100 MHz of this spectrum is also ISM spectrum, with the potential of industrial devices being operated there also. But the other 200 MHz is "virgin" spectrum and reserved exclusively for communications.
A number of companies now offer equipment for the 5 GHz band, and as the price of RF components for 5 GHz continues to fall, more and more equipment will become available. The emergence of equipment compliant with the 802.11a Wireless LAN standard is expected to play a major role in increasing the popularity of equipment for the 5 GHz band. Where 802.11b offers (theoretical) speeds up to 11 Mbps and operates in the 2.4 GHz band, 802.11a offers (again, theoretical) speeds up to 54 Mbps and operates in the 5 GHz band.
If RF Lighting is an "ELE" to communications users of the 2.4 GHz band, at least there is time to begin "planning for survival", planning for migration to 5 GHz, study of new equipment, lining up additional financing, etc.
Steve Stroh
Copyright (c) 1997-2004 by Steve Stroh.
