I'm working on a project to design a space-based low frequency array. It will only be receiving signals, and it will operate between 10MHz - 100MHz. It is made of short wave crossed dipole antennas. I want to get an estimate for the power consumption and data rates, and my web searches have been unhelpful.

When I look for power needs, all I find is regarding transmitters. I know I will need to pass some current through the antennas along transmission lines, but I have no clue how much. This is more of a mechanical engineering project, so at the end of the day I need a wattage value. Is there a general equation that determines power needs, maybe based on wavelength and array geometry?

Second, I need some data rate estimate to size the onboard computing. Given that I'm clearly not an electrical engineer, this may not even be the right question to ask. What determines data output rates in a receiver system? Is it the antenna itself, the source, or the sampling of the computer?

Thanks for any insight.


The antenna itself doesn't require any power because it receives power from the transmitter. But you might need to power a preamplifier. The power consumption of the preamplifier will depend greatly on the design, but for ordinary designs it's very small, perhaps 25 mW. Much less, if you make low power a priority in the design.

The data rate is limited by the Nyquist theorem: you must have at least 2 samples per second for every Hz of bandwidth, assuming you want to sample the signal and process it digitally. Often the sampling is done at a higher rate, then filtered and converted to a lower rate digitally (a process called decimation) because it yields more flexibility and it can reduce the design requirements on the analog filtering which is more difficult to realize. If you're demodulating something digital, often after clock recovery the stream is further decimated to one sample per symbol.

  • $\begingroup$ Thank you. Just to clarify, when you say transmitter, do you mean the radio source? $\endgroup$
    – Spuds
    May 2 '19 at 3:00
  • $\begingroup$ Thanks, corrected. $\endgroup$ May 2 '19 at 12:07
  • $\begingroup$ Assuming any realizable anti-aliasing filter, the sample rate has to be higher the twice the bandwidth, not "at least". $\endgroup$
    – hotpaw2
    May 3 '19 at 18:09
  • $\begingroup$ @hotpaw2 How much higher? 0.001%? 10%? 500%? Discussing theoretical limits is not an unusual thing to do when there's insufficient information to estimate the practical performance of a thing. $\endgroup$ May 3 '19 at 18:57
  • $\begingroup$ All those are non-zero amounts. Zero is incorrect for any non-fictional data. $\endgroup$
    – hotpaw2
    May 3 '19 at 19:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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