Consider an antenna as a 2-port reciprocal network such that you may have S11 reflection at the feedpoint, but also S22 reflection back into freespace.

S11 is well understood and measured because that is usually what you care about for transmission, but this question about S22: impedance mismatch between the antenna and freespace (376Ω) will reflect the signal you want to recieve, back into freespace away from the antenna.

  • Are their known S22 freespace reflection values for different kinds of antennas?
    • Dipole
    • Yagi
    • Helical, ...
  • Preferrably, can S22 be inferred or calculated from S11?
  • If not then can S22 be simulated or measured?
  • Other considerations?
  • \$\begingroup\$ basically, the amount of energy not reflected is (neglecting material losses) the antenna efficiency - and that's linked (by definition of that) to the effective area of the antenna. \$\endgroup\$ Commented Oct 8, 2021 at 22:02
  • \$\begingroup\$ I honestly haven't thought about this, until now. But thinking about it now, I think the idea is to model things such that that freespace is taken as a secondary load whose impedance is "transformed" by the feed line to the primary where that load "is seen." The mismatch is lumped there, not the other way 'round. But perhaps that's just one of many approaches. You might choose to see a great match at one end and push the mismatch to the other end, or distribute it to both ends. Just my guess. Makes me want to play with finite elements to see what comes. Someone must have already done that. \$\endgroup\$
    – jonk
    Commented Oct 9, 2021 at 19:42

1 Answer 1


Whether or a freespace mismatch generally reflects is not defined – imagine this:

you put a high-resistance, very thin cable that's got nothing to do with the wavelength into free space. It's not well-matched to free-space impedance at the frequency you care about at all! But it will not reflect very much; the fact that mismatches reflect is something that follows from the boundary conditions in transmission lines; take these boundary conditions away, and the wavefront might just mostly "ignore" the antenna, it might reflect it, might reflect partially and let through partially (or mix any of this).

  • \$\begingroup\$ I'm still fishing for something to say about S22 for the touchstone file in the previous question. Should I assume S22 is near 0? \$\endgroup\$
    – KJ7LNW
    Commented Oct 8, 2021 at 23:19

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.