-1
\$\begingroup\$

Let’s assume that we have a cavity backed antenna which has half-omni coverage, or call it unidirectional. We know that making it more directional will cause narrow 3-dB beamwidth, so above the horizon gain (such as 30 degree above the horizon) reduces.

What would you try to make the peak gain (assume that pattern is symmetric) remain same but enhancing above the horizon gain?

\$\endgroup\$
8
  • \$\begingroup\$ It depends on your application. Where do you need gain? Where is the antenna pointing? Most antennas are directed with max gain toward the horizon. Are you trying to point into the sky? \$\endgroup\$
    – Alphy13
    Jun 29, 2022 at 19:30
  • \$\begingroup\$ Your assumptions are inconsistent with wave theory \$\endgroup\$ Jun 29, 2022 at 19:43
  • \$\begingroup\$ What you want to do requires (I think) a multi-element antenna (a phased array) with amplitude and phase control of each element so that a complex weighting function can be applied. \$\endgroup\$
    – SteveSh
    Jun 29, 2022 at 19:44
  • 1
    \$\begingroup\$ And draw us a picture of what you want the antenna pattern to look like, in the elevation plane. \$\endgroup\$
    – SteveSh
    Jun 29, 2022 at 19:46
  • \$\begingroup\$ This questions needs a lot more details \$\endgroup\$ Jun 29, 2022 at 19:49

2 Answers 2

0
\$\begingroup\$

The resulting antenna gain is the sum of diversity and mismatch losses.

  • Any focusing of beamwidth increases intensity or relative diversity gain.
  • A spherical isotropic surface defined as gain = 0dBi .
  • For a hemisphere it is 3dBi = gain and for a half-wave ideal dipole, it is 4dBi.
  • But the smaller the focus, the greater sensitivity to tolerance errors and losses.

Consider the equivalent in light intensity. Each 50% reduction in beam angle results in 2x intensity minus losses.

Estimate your focal wave and reflection shape in both horizontal and vertical and figure out where the errors in gain from. diversity, Return Loss or geometric errors or shape errors.

https://en.wikipedia.org/wiki/Dipole_antenna

Perhaps you need to define some specs 1st for diversity gain, or beamwidth, carrier bandwidth, s11 or Return Loss, and then antenna type. Maybe you want a corner reflector to reduce the vertical angle to void Multipath losses from the ground or Rice Losses but also increase width using vertically stacked emitters. You may consider if the receiver expects a polarized signal or not or if multipath is expected and relies on modem correction.

\$\endgroup\$
0
\$\begingroup\$

Gain:

Gain = Efficiency*Directivity

If you add power in one direction, you are taking it from somewhere else. Directivity is just controlling which direction the antenna works best in. You can get very good performance in one small area by ignoring signals in other directions. If you want more power everywhere, then you need to either increase the efficiency or just input more power.

Directivity:

If you want an antenna pattern that has a peak at zenith and at 30deg above the horizon, its doable and you will find several antennas that have a sidelobe in the direction you want, but this is usually treated as undesirable behavior.

If you want to work in the entire area from zenith to 30deg above the horizon, you will not get very high gain by definition. Just aim for a Half Power Beam Width (HPBW) of 120deg.

Assuming you are doing something like tracking a satellite, you can use a phased array to either follow a specific object, or to scan the operating area. This works by actively moving the area with the most directivity.

Efficiency:

You can improve efficiency by making sure you antenna is well tuned to your frequency. (A very efficient antenna is typically very narrow band.)

You can use low loss materials.

You can make sure that the matching of your antenna to the rest of the system is good (this is typically also narrow band).

You can make sure that the polarization matches the opposing antenna.

\$\endgroup\$

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.