measure/calculate antenna isolation betwwen tx/rx antennas 1

[BE123]

hi, I have two antennas - one transmit the other receive. They both work in the same band (GHz).
I know the Radiation pattern of both antennas,
I want to put one antenna above the other, I want to know how to calculate the isolation so, the transmitting antenna wont affect the receiving one.
any idea?


thnx

 

[VEBill]

It's a very complex topic, and your question is a bit open ended. Can you provide more details to help focus on your specific application?

What frequency range - how many GHz?
What type of antenna?
How much spacing between the two?

Be aware that it's probably not worth trying to do much better than an estimate.

Interfering is also a function of the transmiter and receiver.

 

[BE123]

hi,
I am talking about C-band
I am using dish antenna

distance- that what i want to find, I want to chose the isolation and then find the distance that I need.


thanks

 

[VEBill]

Trying to use an estimate to tweak a distance indicates that you've not brought onboard the likely magnitude of the error bounds.

I think that it would be better to estimate approximately how much antenna-to-antenna isolation your transmit and receive equipment requires, then estimate approximately what the coupling will be, then compare the two to see if it's even close. If the two values are far apart, then either it's good news or bad news.

Let me try to explain why the error is likely to be large in this situation.

If you're calculating pathloss between two antennas, when they're aimed towards each other, the antenna gain on the bore sight is well specified. In those circumstances, it's feasible to calculate a reasonably good estimate using the old "32.45" pathloss formula (just Google 32.45 and you'll find it). It'll still be an estimate, and external factors will increase the pathloss on bad days.

When the coupling depends on the two antenna's side lobes, and you intentionally arrange things to try to get into the nulls, then the coupling begins to have sensitive dependence on uncontrolled details. Are you in the Near Field? Is the coupling via diffraction at the edge of the dish? Do the nulls in the side lobes rotate due to a myriad of variables? Are there multiple effects such that there is multipath, thus making the coupling a complex function of frequency?

Have you considered installing just one antenna and using a duplexer to filter the transmit and receive bands? Virtually all real world applications find it cheaper to use a duplexer rather than install and maintain a 2nd antenna. This assumes the T/R frequencies are separate (a mandatory requirement).

To calculate the rough order of magnitude coupling, use the pathloss formula and then adjust for the side lobes. Use the min (nulls) and max (peaks) side lobes to speculate on the error bounds.

Estimate your Near Field distance, tricky because the aperture doesn't really apply to side lobes.

Compare this rough estimate to what your equipment requires. Don't forget to include cable losses in your spreadsheet.

Keep in mind the error bounds. Only an issue if your numbers land in the decision middle.

Consider using a duplexer and just one dish. Look up 'Low PIM cable' for related info if your go this route.

 

[BE123]

the tx and rx at the same freq.

 

[Comcokid]

Why calculate? You might miss some other effect (a back-side lobe reflecting off the mast, etc). Use a network analyzer, or, transmit a constant signal on one antenna, and put the other into a spectrum analyzer to measure the level.

 

[VEBill]

"...tx and rx at the same freq."

That makes it simple.

Answer: There's not enough isolation. Not even close. Probably 60dB +/- 30dB too much coupling.

Disclaimer: there might be some extreme and impractical cases where it could be made to work. If it was practical, everybody would reuse the same frequency for full duplex. Nobody does.