
antennatheory.com Antenna Theory

View previous topic :: View next topic 
Author 
Message 
ronnu AntennaTheory.com Newbie
Joined: 16 Mar 2017 Posts: 3

Posted: Thu Mar 16, 2017 7:54 am Post subject: Antenna aperature 


I have to place two antennas, which both transimit and receive, close to each other. The antennas work on completly different frequencies so there shoudl be no interferences, but still, I would like to assess this.
I would like to keep it as simple as possible so I would simply assess the isolation between those two antennas. I understand that I can use friis formula for that, but what should be the wavelength used for calculation? I understand that it's dependent on antenna aperture which is incase dependent on freqcuency/wavelength.
For example, I have 1090 MHz (antenna 1) and 433 MHz (antenna 2) antennas. I would like to assess the isolation from antenna 1 to antenna 2, but what should be the wavelength used for calculation? I think it should be 433MHz wavelength, but would like to verify it. 

Back to top 


coundpake Antenna Theory Regular
Joined: 03 Aug 2016 Posts: 10 Location: San Jose, CA

Posted: Thu Mar 16, 2017 7:23 pm Post subject: 


Best way to assess this would be to use a network analyzer to measure S21 between the two antennas, or if you dont have that, a signal generator and power meter.
But if you want to approximate the isolation..
1) Friis equation approximates free space path loss at a single frequency. I think you should assume a frequency between 1090 MHz and 433 MHz (maybe the mean, 761 MHz)
2) Friis equation does not work in the nearfield. For 433 MHz, they need to be at least 2 wavelengths or 1.5m away from each other.
3) Antennas should be in line of sight of each other with no multipath
4) You will need the gain of each antenna (Gt, Gr) in the direction of the other or you can assume G=1 for omnidirectional (which wont be a horrible assumption for dipole).
5) You also need to consider polarization mismatch loss. In the case of dipoles, this will be significant if they are not parallel to each other.
6) there are also a few other less significant factors.. check out the Friis Transmission Equation Wikipedia
TLDR: You can use Friis to approximate the isolation but it will always have error. Make the measurement with VNA or calculate by making some assumptions but leave yourself plenty of margin for the error. 

Back to top 


ronnu AntennaTheory.com Newbie
Joined: 16 Mar 2017 Posts: 3

Posted: Sat Mar 18, 2017 4:36 pm Post subject: 


Thanks for the reply coundpake.
Yeap, eventually I will measure the isolation between antennas, but first I would like to get a rough estimation. That would let me design a inital placement of antennas which I would then test.
I made some calculations and using the frequency of 761 MHz I got isolation around 30dBm. But that doesn't seem right based on experience, because systems which work on frequencies so far apart should definelty have more isolation than that.
It seems to me that this isn't the right approach, but I think that maybe I have to take into account the gain pattern of the antenna at the specified frequency. Antenna radiation pattern is dependent on the frequency and I did some simulations with 4NEC2 program and it showed that with longer antennas (relative to wavelength) the radiaation pattern changes completly and 433MHz antenna working with 1090MHz signal would have it's maximum gain (1.2dBi) pointed 45 degrees up from the xaxis and would be around 46dB in the direction of xaxis. In addition to this there would be path loss which in total would give around 75dB of isolation. That seems more realistic to me. Any ideas if that would be correct assumption?
The antennas will be both vertically polarized and in LOS . Maximum distance that they can have from each other is maybe a meter. You're right that Friis formula cannot be use in nearfield but I found that for 433 MHz tuned monopole should have a boundary to farfield around 0.08m. I used different calculators found in the web:
https://www.everythingrf.com/rfcalculators/antennanearfielddistancecalculator
http://www.rfwirelessworld.com/calculators/antennanearfielddistancecalculator.htm
Both gave same results. Maybe there is some literature that covers this thing in more detail? I couldn't find anything in the net, but maybe someone knows a book or some article? 

Back to top 


ronnu AntennaTheory.com Newbie
Joined: 16 Mar 2017 Posts: 3

Posted: Wed Mar 22, 2017 7:00 pm Post subject: 


Okay, so I kept digging and this is what I've come up so far. The antenna gain is in fact dependent on the frequency as well is antenna efficency, which both will degrade if antenna is not operating on the frequency it was designed for. Main thing that affects isolation between antennas is distance and I'v used the frequency of the transmitting system for calculations in Friis formula as this is what the receiver is receiving (don't know how I didn't see the logic before). The isolation between antenna might not be that much, especially when their close to each other. The main attenuation comes from the receiver/transmitter itself. Both have filters which attenuate any outofband frequencies. Again, don't know how I didn't see the logic before.
To evaluate the interoperability with other radio systems you basically have to find what is the signal level transmitter generates in the input of the receiver (taking into account all the losses and gains). Then find out what is receivers sensitivity and minimum SNR and subract former from the latter which will give you noise floor. If the received interference signal is higher than noise floor, then sensitivity of the receiver will degrade bacause receiver will see that interfereing signal as noise and as such it will raise the noise floor.
Hopefully it will be of help to somebody in the future! 

Back to top 




You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum

Powered by phpBB © 2001, 2005 phpBB Group
