Engineers are an amazing bunch. Every time the establishment proclaims that something cannot be done, engineers (often flanked by their friends in science and math fields) seem to figure out how to get it done.
When it comes to the vast electromagnetic spectrum, this has been a mind-boggling last few decades in which engineers have squeezed more and more bandwidth out of scarce resource - all while tweaking modulation schemes and sending information over parts of the spectrum once thought unusable for communications.
In particular, the last few years have pushed the boundaries as consumers and businesses clamor for bandwidth. The culture of constant wireless connectivity (whether truly desired or simply the result of societal pressure) has only grown in tandem with the explosion of the Internet.
And yet, with every year, it seems there is less and less spectrum available for commercial use. Some of the best spectrum, after all, is designated for the militaries and governments of the world. Other high-quality bandwidth spreads are used for broadcast television and other one-way media. So with incredible scarcity and the public's insatiable demand for wireless services intersecting like never before, the next century may belong to the engineers. These are the technological magicians who will twist, crunch, compress, and amplify - even tame - wireless frequencies in ways never before thought possible.
"There is no limit to the spectrum because you can just keep reusing it," says Peter Cochrane, a technical advisory board member to e-tenna, which designs more efficient receiver and transmission antennas for wireless networks. "You can just keep climbing to 80, 120, 180 gigahertz. It's all modulation."
Of course, all of this begs the question: What are the limits of physics and economics? How much spectrum is really available for communications services? Cochrane notes that trying to send information in the upper microwave band and beyond has its challenges, which increase as the frequency gets higher.
"You start to resonate with the molecules in the atmosphere, so the energy dies very quickly," he says. "It's like something the size of a pea trying to make its way through a field of soccer balls."
For their part, e-tenna engineers are working on new antennas that can hone in on specific cell towers to make more efficient use of bandwidth - all without moving parts. The same technology can be used to allow the cell towers themselves to find other towers without human beings physically "pointing" them at each other. "Antennas can then adapt spacially and promote very high data rates," says Andy Humen, e-tenna's co-founder and director of marketing.
Engineers seem to relish the neverending challenge of eking out bandwidth where it didn't seem to exist yesterday or pushing the limits of usable spectrum. After all, in recent years they have figured out how to modulate frequencies in the 40-gigahertz range in ways that could allow for feasible telecom services. In fact, the Federal Communications Commission has licensed several companies in that band, which is known as local multipoint distribution service (LMDS). The licensees are still building out their networks, however, so it's unclear whether successes in the lab will translate into successful business strategies.
Still, considering the history of spectrum exploitation, you have to wonder whether engineers will someday make use of frequencies even higher in the spectrum such as X-rays, gamma rays and even cosmic rays for communications.
It seems a stretch for various reasons, not the least of which is that the potentially unsafe radiation levels get hard to tame in the ultraviolet and higher frequencies. "You certainly don't want to sunburn people," jokes Richard Caldwell, dean of the school of natural sciences and mathematics at the University of Texas.
High frequencies not only fall prey to "absorption" by air molecules (the pea-and-soccer-ball problem), but tiny wavelengths can also bow to everyday obstacles such as trees, buildings, and even rainfall. High frequencies also travel extremely short distances, requiring massive infrastructure to deploy a service throughout a local area.
As a basic rule, the higher the frequency, the greater number of base stations needed to push the signal out to wider areas. The tradeoff is that high frequencies can tout higher bitrates than lower frequencies. But these are general guidelines. Signal performance also has much to do with the algorithmic properties of the signals themselves. "It all depends on how effectively you could modulate the signals," says Caldwell. "Today, there are some practical limitations, but who knows what might happen in 100 years?"
Indeed, the far-out future is almost impossible to predict. It seems unlikely that scientists in the 1800s could have imagined radio frequencies becoming the high-speed carriers of digital information over vast distances, in and out of small handheld devices. That would have required a leap of unreasonable proportions, not to mention knowledge of future inventions such as the transistor and, more importantly, the microchip.
Could it be just as impossible for contemporaries of the early 21st century to imagine modulation techniques - perhaps using new power schemes or unimaginable computer computation skills - to safely use parts of the spectrum now thought unusable? Only time will tell.
To truly understand the limits of spectrum use, it's important to understand the most basic properties of the spectrum itself. The electromagnetic spectrum is light, only a small portion of which is visible to the naked, human eye. The rest consists of everything from radio waves, in which the distance between the crest and trough can be several kilometers, to gamma rays and cosmic rays, whose wavelengths are microscopic.
Because the spectrum travels at the speed of light, it can be measured by the number of times per second that a wave passes a given point. One wave cycle per second is one "hertz." So lower frequencies with longer wavelengths are designated with a lower hertz than their high-frequency cousins.
The various frequencies used for VHF broadcast TV signals, for example, emit between 54 to 216 million hertz, or "megahertz," per second. Microwaves, which are used for most commercial wireless services such as PCS, MMDS, LMDS, etc., generally fall between 100 MHz and 100 GHz. Gamma and cosmic rays emit so many waves per second, it's difficult to even describe it (1019 hertz to at least 1025 hertz, to be exact).
The anatomy of mobile data
As a general rule, the lower radio and microwave frequencies are used for telecommunications because they travel farther and penetrate obstacles much more easily - key factors when deploying a commercial service across geography that can span several miles in diameter. Aside from looking at bone fractures or fueling comic books about green superheroes, X-rays and gamma rays have far fewer commercial applications under the constraints of today's technology.
The trade-off between frequency and distance has much to do with the current allocation of spectrum throughout the modern world. Television and radio signals, for example, travel across relatively low frequencies (under 1 GHz), which is why it's often easy to receive those signals from a TV antenna in the basement.
A notch above the lower radio spectrum lies most of the mobile wireless bands used by businesses and consumers. Because such microwave signals don't penetrate structures as easily, it's much harder to receive a PCS signal in a basement. Elevators, which encase users in a cubical of signal-reflective metal, are even more difficult for PCS signals to bore through.
On the bottom end of the spectrum are much lower frequencies, whose wavelengths can be enormous (at 10-KHz, for example, waves are about 30 kilometers long!). Military submarines often use such frequencies to send short messages over vast distances (a similar principle is used by whales, only they obviously use infrasonic sound waves rather than low-frequency light spectrum emissions).
But while long wavelengths can go on and on for miles, the infrequency of the crests and troughs greatly limits the potential bitrates. Short of using laser technology or other new techniques, forget about broadband speeds at these low frequencies. According to Cochrane, these lower bands churn out as little as one bit per second in exchange for the ability to easily traverse great distances.
So commercial applications of the future may have to push the boundaries of physics to eek out higher bitrates, penetration and distances than currently available. Of course, we should all take heart: No one ever thought fiber optics would yield the mind-blowing capacity since made possible by wave-division multiplexing, which involves splitting single beams of light into multiple beams able to carry more streams of data (without adding any new fiber).
If photonics can yield such advances, who knows what's in store for the seemingly finite electromagnetic spectrum?
"Spectrum is fundamentally unlimited if you're willing to put the money into it," points out David Crosbie, CTO of Boston-based Bluesocket. And while those factors are up to the financiers, the world should feel confident that the engineers will handle the rest.
History suggests that they wouldn't have it any other way.
Michael Grebb has previously written for The Industry Standard, Business 2.0, and eCompany. From Washington DC, he covers the impact of mobile technology on modern society.