Using the Bigger Shoehorn!
By The PHANTOM –
Last quarter we talked about the bigger shoehorn of coherent fiber optics as one way to get more out of fiber we already have. You’ll recall (or maybe won’t recall) the simple definition my small brain uses, that we are treating fiber optics more like the way we treat RF-based signals. Rather than simple off-on modulation of the light (akin to the century old RF technique of code transmission), we can change the light carrier in ways that let us get more information through.
The magic is in getting lasers that put out coherent light (as you might expect from the name—duhh). Normal lasers put out short bursts of energy then shift to a slightly different frequency (or wavelength—the two terms are inversely related). Any significant shift blows to smithereens the idea of treating the laser beam like an RF signal. So what does “significant shift” mean? Basically it means that the shift must be low enough that we can still detect the output. As to the details, we leave that up to the lab geniuses who look at bit length and other parameters to decide.
So, what types of modulation formats might we be looking at near-term, and what are some of the detection (demodulation) schemes around? Well, pretty much where we were with RF a number of decades ago. Simple modulation schemes include on-off keying of the light. No, that’s not a coherent technique, but we sometimes start here for backwards-compatibility, a necessary thing in any industry that wants to move ahead without breaking the bank. We can move on to four level amplitude modulation, not really a coherent technique in my comic book, but a next step, in which we use four distinct amplitude levels of light. What we’re really doing here is to take a baby step back toward the analog domain in the interest of improving bandwidth efficiency. Improving bandwidth efficiency is a fancy way of saying that we want to get more speed for the same RF (or optical) bandwidth. All else being the same, if you want to get more data (bits per second) through, you have to use more bandwidth, and bandwidth ain’t an infinite resource. So to get more bits per second for no more bandwidth, you gotta do something with modulation. For example, in order to get more bandwidth efficiency, I can go from two-level (on-off) modulation to four level. With two-level, maybe full carrier corresponds to a 1 being transmitted and no carrier to a 0. But four level means I transmit, say, 0 level for the two bits being 00, 33% level for 01, 67% level for 10 and 100% level for 11. So now during each bit period I transmit something representing two bits, doubling my throughput without widening the occupied bandwidth.
While I can do it this way, it turns out that a better way is to vary the phase of the RF (or optical) carrier; maybe 0 phase (I have to have a way to define 0 phase, but let’s not go there now) corresponds to 00, +90 degrees to 01, 180 degrees to 10 and 270 degrees (a.k.a., -90 degrees) to 11. Now we’re getting into the place where we need to talk about coherent optics for transmission: If you’re gonna talk phase shift we have to use a coherent scheme whereby we have a consistent phase reference to talk about.
Those of you who have not dozed off by now recognize this as a simplified description of quadrature (for 4) phase shift keying, or QPSK. We were there in practical deployments in the early ’90s for RF—early data transmission schemes for voice and data (before the two merged into one) used it. In the optical domain when we talk about a particular phase we must be able to expect the phase to not wander off for long enough to get the bits through. We can use any of a few techniques to do the detection of the signal. One, coming straight from the RF world, is to heterodyne the optical signal down to where we can convert to RF in an optical detector (as we do today), and get out an RF signal of some convenient frequency, say 1 GHz or lower. We do this by running a laser local oscillator 1 GHz off in frequency from the frequency of the transmit laser, combing the two and going into the optical detector. Pretty straightforward assuming you have coherent lasers at both the transmit and receive ends. There are other techniques whereby we operate the receive laser as part of the detection process itself. All of this is in a pretty rudimentary state right now, as we were with RF modulation decades ago, but the future is bright!
You never know when The “masked” Phantom is standing right beside you. Sometimes he is in a Zoom meeting or virtual SCTE training session with you. He may be hanging with the suits and other times with the front liners. But be assured, The Phantom knows all and, most importantly, The Phantom sees all!
—
Shutterstock
