The More You Know

In the spring of 1983, the Radio Netherlands program Media Network featured an interview with John Branegan, GM4IHJ. He was talking about a constellation of amateur radio satellites that had CW beacons on ten meters. The beacons transmitted the satellite's ID, along with some telemetry data. The point of the interview was to generate interest among SWLs for monitoring and tracking low earth orbit satellites. It worked. I immediately wrote to Radio Netherlands to request a copy of Branegan's pamphlet Satellites for the Short Wave Listener. By the summer of '83—the summer between my freshman and sophomore years of high school and the last summer that I wouldn't have a job—I was actively tracking and predicting orbital passes of a half dozen satellites using nothing but pen, paper, a calculator, and my trusty Radio Shack DX-100.

When I would tune in a satellite, the audible CW tone coming out my speaker would glissando up throughout the pass, unless I made continuous adjustments to the BFO or tuning knob. The satellite's transmitting frequency, as seen from my listening position, was continuously dropping, increasing the gap between the satellite's frequency that of the BFO, thereby increasing the frequency of the heterodyne between the two. The drop in apparent frequency was due to the Doppler shift. This came as no surprise; it was described in Branegan's pamphlet and was something I intuitively understood from hearing the sound of a passing train's horn. 

When I earned my amateur radio license many years later, it seemed only natural that I'd want to return to satellites. Based on my earlier experience, making Doppler shift adjustments to either my transmit or receive frequency was second nature, and achieving early success as a satellite operator was relatively easy.

Thinking about what I was doing every time I bumped that frequency knob led to a bit deeper thinking, and a couple questions. Often when a random technical question comes to mind, I turn to ham radio Twitter as a kind of Elmer-by-committee, typically prefacing these naive questions with a title like Today's Question From a Guy Who Went to a Liberal Arts School. Ham radio Twitter has responded generously, and I am grateful to the whole community for all that I've learned.

One such question had to do with the modulated signal. The radio frequency signal is subject to the Doppler shift, why not the modulated signal? When I'm making satellite contacts, why don't I hear the other operator's voice dropping in pitch as though he or she is shouting out the window of an automobile that's doing 28,000 kilometers per hour? The reason has to do with the constancy of the speed of light. The speed of light is constant, regardless of whether the source and observer are moving toward each other, moving away from each other, or maintaining their relative positions. Physicists like to say it this way: the speed of light is the same in all frames of reference. This isn't just some accepted first principle, it's supported by experimental observations. The satellite's radio wavelength gets compressed or stretched depending on whether the satellite is moving toward or away from my receiver, but the signal itself—and with it the modulated information—always gets to my antenna at the same 300,000 kilometers per second. Another way to picture this is by using the wave/particle duality of light. Light (and radio, which is just lower frequency light) is both a wave and a particle. Or, to make visualization simpler, it's a stream of particles traveling from transmitter antenna to receiver antenna. We can think of each individual light particle, or photon, as a snapshot of the state of the modulated signal at a point in time. My receiver reconstructs the audio frequency waveform of the other operator from a series of snapshots that arrive at a constant speed, no matter how fast or in what direction the source is moving.

Does this mean there's nothing weird about what the Doppler effect does to a radio signal? Hardly. A more perplexing question popped up as I was thinking over this wave/particle duality. Satellite AO-91, to use as an example, receives on 435.250 MHz. At AOS, I tune my transmitter to 435.240 MHz so that the satellite, approaching me at 28,000 kilometers per hour, will hear me on a Doppler-shifted frequency of 435.250 MHz, or close enough to it that the FM receiver can lock on. Every few minutes, I adjust my frequency up 5 kHz so that just before LOS I'm transmitting on 435.260 MHz and the satellite, now moving away from me, will still hear my signal, Doppler shifted down to near the same 435.250 MHz. Other than at the exact moment of the satellite's closest point of approach to my QTH, the satellite is receiving me on a slightly different frequency than I'm transmitting on. Thinking of radio as a particle, the energy conveyed by each photon is directly proportional to its frequency, expressed by the formula E = hf, where E is the energy of an individual photon, f is frequency and h is Planck's constant. When I transmit, I'm sending a stream of photons up to the satellite. I could, in theory, predict exactly how many of those photons will strike the satellite's receiving antenna. From the satellite's perspective, its receiving antenna will indeed be struck by the predicted number of photons, but because of the frequency shift, each photon will be at a slightly different energy level than what I sent. The satellite and I will disagree on the amount of energy being exchanged between my transmitter and its receiver. Doesn't that break conservation? The answer is no, the disagreement in energy exchanged is real, but it is predictable and doesn't break conservation, because conservation has to be thought of differently under these conditions. Special relativity is a consequence of the speed of light being constant in all frames of reference. Observers in different frames of reference will disagree on the simultaneity of spatially separated events in ways that can be predicted. It turns out that different frames of reference will also disagree on the amount of energy they exchange; energy is only conserved within a single frame of reference.

I'll leave with one last, albeit trivial, observation about the speed of light. Last weekend I worked French station F4ILH while on a POTA activation at Dockweiler State Beach, off the departure end of runway 25R at Los Angeles International Airport. The station was just over 9000 kilometers from my QTH. At the speed of light, it took my signal a little over 30 milliseconds to reach him, and his signal just over 30 ms to reach me. I find that incredible for two entirely contradictory reasons. First, it's incredible because a distance that once would once have been a months-long sailing voyage and today is an eleven-hour plane trip can be traversed by the signal from my modest HF rig in a small fraction of a second. But it's also interesting to think of the time gap—latency in networking parlance—involved. Thirty milliseconds is a short, but nontrivial, span of time. If you heard two pulses spaced thirty milliseconds apart, you would hear the two pulses distinctly and would be aware of the gap in between the two. It's a bit awe-inspiring to think that on that on that cool autumn morning on a California beach I was making contact with a station so far away that even at the speed of light, the speed of causality itself, it took a perceptible amount of time for my signal to reach him. I'm really going to have to try EME one of these days.