From Scientific American, March 2021:

Optics. Traveling Photons. Joanna Thompson.

Much of today's space communication relies on radio signals. But these diffract and broaden as they travel, as does light or any other electromagnetic wave. A radio beam fired from the moon to Earth “would typically diverge to the size of a continent,” says Peter Andrekson, a photonics researcher at Chalmers University of Technology in Sweden and co-author of a new study in Light: Science and Applications. In contrast, he notes, “a laser beam would diverge to a two-kilometer radius or so.”

Catching enough of a spacefaring radio signal from somewhere like Mars requires a really big dish. NASA's widest receivers stretch 70 meters across, says Bryan Robinson, an optical communications engineer at the MIT Lincoln Laboratory, who was not involved in the study: “It's like a football field that's on a gimbal pointing to Mars.”

The rovers and probes out there in space don't have a 70-meter dish... So how do they receive OUR signals?

And if Mars is distant enough to make communication so difficult, how do we send/receive signals to and from the New Horizons probe, out past Pluto?

P.S.: When mirrors on telescopes, as well as radio dishes, are being officially 'measured', do they measure in a straight line from edge to edge or do they lay the tape measure along the curve of the hyperbolic or parabolic dish?


1 Answer 1


We communicate via radio, as you note, but it's not quite as hard as your question makes it sound. Don't get me wrong, it's still an amazing feat of engineering, but we can walk through how it's possible.

First, we can think of this by way of an analogy. You might think that this is like being in a stadium at night, with huge, bright, white lights shining down from on high, and you're trying to pick out someone with a tiny pen-size flashlight who's shining it from right next to those big lights. But, that's not a good analogy. Instead, we use radio wavelengths that are relatively quiet so far as most other signals from the cosmos are concerned. So, it's more like you have those really bright stadium lights that only emit light from blue thru red, and you're shining a small ultraviolet light from it. If you have an ultraviolet filter, those bright stadium lights will effectively not even be on, and you can easily pick out the tiny ultraviolet light shining next to them.

Second, we have the issue of signal strength. When NASA (or ESA, or another space agency) transmits a signal that they want to picked up by a spacecraft, that signal is extremely powerful. The antennae can transmit a signal with up to roughly 20 kW (see, for example, this other StackExchange question). Enough power to run a large house, all concentrated into one radio signal at a very, very specific frequency. These transmitters are exceedingly good, varying by less than 1 part in 1,000. It's similar to why a 5 milliWatt green laser pointer can look so bright, but a 100 Watt light bulb is a comfortable reading light: All that power is concentrated in just that specific frequency (this isn't a perfect analogy, since it also has to do with how your eye focuses light).

Third, how much information can be transmitted per unit of time (like per second) is then based on how much you can divide that power up and still have it received by the spacecraft. If you divide that 20 kW signal up into 20 pieces per second, you're sending 20 bits per second, and each bit has 1 kW of power in it upon transmission.

Fourth, the light certainly spreads out and attenuates as it travels, following the inverse-square law, and so we have to design the receiver antennae on the spacecraft to be able to detect those small signals above the level of background noise. If the spacecraft is farther from Earth, we have to decrease the bit rate so that each bit has enough power in it to be detected. The received power is often expressed in dB (decibels), like sound. I don't have numbers for Mars craft, but I can tell you for New Horizons, at Pluto, with a 10 kW signal, the antenna would get a signal of about 50–55 dB, which is well above the background noise from any other sources. If it were sound, that's a normal household conversation level of sound.

Fifth is getting a signal from the spacecraft to Earth. Perhaps obviously, the transmitters on these spacecraft don't have 10s of kW to put into their signal. Instead, they have to rely on two things: Lower the bit rate so that each bit has more power in it, and rely on the Deep Space Network telescopes being bigger and so able to detect weaker signals than those spacecraft can. The Deep Space Network 70-m dishes themselves can detect signals in the pW range (picoWatts, or 10-12 Watts) before it gets lost in the noise, provided that that signal is similarly transmitted at a wavelength that other random cosmic sources are not loud on. If I'm doing my conversion correctly, 1 pW is -90 dBm or -120 dBW (dB referenced to 1 mW or 1 W of power).

The math is then reasonably straightforward to say: (a) You can transmit with a certain total power, (b) you can receive signals of a certain minimum power, (c) that total power is going to drop by a certain factor based on how far away you are, (d) so then how much can I divide that dropped power by in order to get my bit rate?

This is why New Horizons when it was near Pluto was transmitting data at around 1300–2400 bits per second, but right now, the bit rate is closer to about 600–700 bits per second, as the craft is about 5 AU farther away and it has less power to put into its transmitter since its Plutonium power source has continued to decay.

Specifically with rovers, those dishes are often smaller and rely on orbiting spacecraft with larger dishes to receive signals from Earth and transmit signals to Earth, effectively as a relay between Earth and the rover.

And, telescope diameter is what's used, so edge-to-edge "as the crow flies," as opposed to how the ant walks.

P.S. I helped design this infographic / information pamphlet for New Horizons that covers a lot of this question.

  • 1
    $\begingroup$ You make a good point, since the gains of both antennas factor in to communications the same way, the thing that determines which direction needs to be at a lower speed comes down to transmitter power and receiver sensitivity. DSN transmitters are a factor of 10${}^3$ more powerful than the spacecraft transmitters, but but DSN receivers can be 30 times colder (10 K versus ~270 K) so thermal noise power per Hz is 30 times lower, which doesn't completely compensate the thousandfold more power, but it helps. $\endgroup$
    – uhoh
    Mar 14, 2021 at 10:51
  • $\begingroup$ Thanks for the long detailed answer, Stuart Robbins! Much appreciated! $\endgroup$
    – Kurt Hikes
    Jan 1, 2022 at 1:32
  • $\begingroup$ P.S. Do you have a Bad Science blog where you talk about b.s. in science and tech, Stuart Robbins? I love reading about that stuff... One of the famous 'experts' on 'bad science', Robert Park at Univ. of Maryland, just died... I loved his 'Voodoo Science' book.... $\endgroup$
    – Kurt Hikes
    Jan 1, 2022 at 1:35
  • 1
    $\begingroup$ @KurtHikes Yes, search the internet for "Exposing PseudoAstronomy." I blogged and did a podcast for several years, but work and emotional exhaustion stopped me about 5 years ago. It's still technically on hiatus as I do want to restart it, but I'm too over-committed with other things right now. $\endgroup$ Jan 1, 2022 at 20:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.