Editor's note: In the first of a three-part series, analyst Craig Mathias details the science behind the technology that underscores today's wireless networks.
Even as someone who's been working in wireless for more than two decades, I still marvel at the technology. Think of it -- we have information in one location, and it appears at the desired destination with no apparent physical connection between the two endpoints. From my first crystal radio set in grade school to the multi-gigabit systems I work on and with today, the magical part of wireless remains.
But, of course, it's not really magic that's behind wireless technology -- it's actually a ton of sophisticated engineering built on some fairly basic principles of physics. So, over the course of three articles, I'm going explain how wireless works. And I promise to keep this simple enough so that even non-science majors will feel the magic at work.
From my first crystal radio set in grade school to the multi-gigabit systems I work on today, the magic in wireless remains.
Let's start at the very beginning. There's a property of the known universe that we call the "electromagnetic spectrum." This refers to the carrier for electromagnetic waves, which we can generate with a relatively simple electronic device called an "oscillator." The oscillator emits sine waves (remember those from high school geometry class?), and we can tune the oscillator so that it generates waves at a specific frequency, referring to how fast the wave we generate vibrates, and amplitude, which is how powerful the waves are. Both of these are usually limited by regulation -- for example, in the United States, the Federal Communications Commission determines who can use what spans of frequencies (called "bands") of the electromagnetic spectrum for what purposes and under what circumstances. As you might guess, the legal issues here are often more complex than all the underlying physics and engineering.
Once we have our sine wave, the next task is to encode upon it the information we wish to transmit. This is called "modulation," and this is performed by changing the physical characteristics of the wave. We can change amplitude (you know this as AM radio), the frequency (FM radio), or the phase of wave, which is where it is at any given moment as we cycle through time across all 360 degrees. And we can combine these variables. For example, quadrature amplitude modulation -- used in satellite communications, modern Wi-Fi systems and even cellular systems like LTE -- is a combination of amplitude and phase modulation. The cleverer we are with modulation, the more data bits can be crammed into the wave. This results in a form of data compression, and thus higher performance, via what we call "spectral efficiency." The last step required to send this modulated signal is to amplify it (add power) and send it out into the world via an antenna. Much more on antennas -- the most important part of the radio -- in the next installment of this series.
How wireless works depends on other factors, too
Radio isn't as straightforward as sine waves and the electromagnetic spectrum. The core problem is that once we transmit the signal, it enters what is known as the "radio channel" -- that property of the universe which allows an electromagnetic wave to propagate from point A to point B. This is where things get particularly ugly. First, radio waves fade exponentially in power ("flat fading") with the square of distance -- which means that even high-power signals get very weak very fast. Thus, the receiver on the other end must be as sensitive as possible to detect the signal, assuming it arrives and is loud enough to be detected. If the signal is too weak, it looks like noise. The goal is a signal-to-noise ratio that's as high as possible.
The world of wireless
Keeping wired, wireless networks in synch
Next, radio waves may be blocked by solid objects ("shadow fading"), or by echoes and reflections of the primary signal ("multipath fading," or "Rayleigh fading"), or by intentional ("jamming," which is rare in the non-military world) or unintentional interference. Wi-Fi and other systems operating in the shared unlicensed bands must use a wide variety of techniques to avoid being clobbered by other signals simultaneously (and legitimately) operating in the unlicensed bands. Not only that, those systems must avoid interfering with signals deemed by regulators to be more important. The most common technique to reduce interference here is the use of various forms of spread-spectrum radio, which literally spreads the signal across a wide range of frequencies in a fashion that improves reliability at the expense of spectral efficiency.
The best is yet to come
But if we're lucky -- and radio does in fact behave quite statistically in most applications -- the signal reaches the desired receiver, where it is amplified, demodulated and converted back into its original form. Most wireless communications today is digital, meaning that we're only sending 1s and 0s, thus making all manner of reliability and performance improvements possible and relatively straightforward. This at least in part explains why today's wireless systems can be so low in cost while being so high in performance -- the 1.3 Gbps throughput of current 802.11ac wireless LAN products, for example, seems like a lot, but even higher speeds are possible and on the way. And all of this is thanks to our ability to design and build reliable, low-cost digital radio systems based upon a few simple laws of physics.