Maarten Baert's website

Game Maker / C++ projects

Home Model Creator ExtremePhysics Game Maker DLLs SimpleScreenRecorder AlterPCB Quadcopters   Recent comments Search

Last modified: Sun, 19 Feb 2017
Refresh

Polarization


These days, pretty much everyone who is serious about FPV is using circularly polarized antennas rather than the much cheaper linearly polarized antennas. You often still get a 'free' linearly polarized antenna when you buy an RX or TX, but I haven't seen anyone actually using those - and for good reason.

But what exactly is the difference between linear and circular polarization? In order to explain this, you need to know a bit more about how electromagnetic waves work. An electromagnetic wave is conceptually similar to waves in water or air, but there is an important difference. Electromagnetic waves consist of an electric field and a magnetic field that interact with each other in a specific way. These field have a magnitude which determines the amount of power transmitted by the wave (or in other words, the signal strength). However unlike waves in water or air, these fields also have a second property: the direction of the field. The electric and magnetic field directions are always perpendicular, and they are both perpendicular to the direction of propagation.

Linear polarization

The following animation shows a vertically polarized wave (which is one type of linear polarization), with blue arrows for the electric field and green arrows for the magnetic field:

Since the electric and magnetic field are perpendicular by definition, it is a common convention to describe only the direction of the electric field. We say that a wave is vertically polarized when the electric field is vertical, which implies that the magnetic field is horizontal. If we just look at the electric field, the wave looks like this:

Now it's easier to see why this is called a vertically polarized wave.

Circular polarization

What does a circularly polarized wave look like? It looks like this:

This is called a right-hand circularly polarized (RHCP) wave. It's called like that because if you make a 'thumbs up' gesture with your right hand, and then rotate it so that your thumb is pointing in the direction that the wave is propagating (i.e. from TX to RX), then your fingers are curved in the same direction as the rotating field. A left-hand circularly polarized (LHCP) wave is the exact opposite, it rotates in the opposite direction. Circularly polarized waves are actually composed of two separate waves, one that is horizontally polarized and one that is vertically polarized. The following animation shows the same wave but with the horizontal and vertical components also shown separately:

So what makes circular polarization so much better for FPV? There are two main reasons:

Reason 1

Circularly polarized antennas will work under any orientation: you can turn them 90 degrees and they still do the exact same thing. This is not the case for linear polarization: if you take an antenna that transmits a vertically polarized wave, and you rotate it 90 degrees, you get a horizontally polarized wave. If you then try to receive this wave with a vertically polarized antenna which isn't rotated, then you won't receive any signal at all. What this means in practice is that when you do a flip or a roll with a RC model and you are using linearly polarized antennas, you risk losing the signal because the antenna is now rotated. And even if you never do complete flips or rolls, this may still affect you, because even a roll of 45 degrees while making a turn will cut the signal strength in half.

Reason 2

Circularly polarized antennas can reduce multipath interference. Multipath interference happens when there is more than one signal path from the TX to the RX. The most common case is a direct path through the air as well as a reflected path against the ground. You might think that having more than one signal path is a good thing, because it allows you to receive more signal power, but that's not how it actually works. Since the two paths have a different length, the two signals will arrive at slightly different times, and this can lead to destructive interference where one wave cancels out the other wave, and the result is no signal at all.

What does all this have to do with circular polarization? Well, when a circularly polarized wave is reflected by something, it acts like a mirror and it switches the polarization around. An RHCP wave becomes an LHCP wave and vice versa. Since an RHCP antenna will only receive RHCP waves, and ignore LHCP waves, the reflected wave should not affect the signal strength at all.

It is often said that reflections will reverse the polarization perfectly, but this is actually an oversimplification. In reality this only happens when the incoming wave is perpendicular to the reflecting surface. If the wave is instead reflected at a shallow angle, then the polarization won't be reversed completely, and you will get a mixture of LHCP and RHCP waves. On top of that it is also possible to get multiple reflections - a signal could first reflect off the ground, then off a building, and then reach the receiver. In this case the polarization is reversed twice, which has no effect. So even with perfect circularly polarized antennas there would still be some multipath interference.

The reality

In reality, there is no such thing as a 'pure' LHCP or RHCP antenna. For practical reasons, any circularly polarized antenna will also unintentionally emit a weak wave with the wrong polarization. These two waves, when added together, create a wave which is shaped like an ellipse:

In this case you could also say that the horizontal and vertical components of the wave don't have the same amplitude. This is an equally valid way to look at it, because mathematically it's the same thing, as you can see in this animation:

Amplitude difference isn't the only possible problem. If the phase difference between the horizontal and vertical component isn't exactly 90 degrees, then you also get an ellipse, just at a different angle:

Axial ratio

The axial ratio of an antenna is a number that is used to describe how close the waves produced by the antenna are to a perfect circle. The axial ratio is equal to the widest part of the ellipse divided by the narrowest part. For a perfect circle, the axial ratio is equal to one. In reality, all antennas will have an axial ratio somewhat greater than one, and the goal is just to get as close as possible.

The reason why people care about the axial ratio is because the two advantages of circular polarization depend on it. For example, an RHCP antenna which actually emits 90% RHCP and 10% LHCP won't completely reject multipath interference, instead it will only reduce the signal power of the incorrect polarization to 10% of the original power. This is still very useful, but not quite as effective as it could be in theory.

Mixing antennas

What would happen when you transmit your signal with a certain type of antenna, and then use the wrong type of antenna to receive it? It depends on the particular combination that you are using:

  • Vertical + horizontal: No signal, in theory. In practice there will usually still be a very weak signal because the antennas are not perfect. It also depends on the orientation of the antennas.

  • LHCP + RHCP: No signal, in theory. In practice you will still get a weak signal because the axial ratio is never perfect. The strength of the signal is dominated by the worst antenna, so just having one perfect antenna is not enough. Both antennas need to have a perfect axial ratio to block the signal completely.

  • Circular (LHCP/RHCP) + linear (vertical/horizontal): You will still receive a signal, but only at 50% of the actual power level. There is also no rejection of LHCP or RHCP, both are received equally well by a linearly polarized antenna, which means this configuration is susceptible to multipath interference. The orientation does not matter though.

As you can see there's really no good reason to mix different types of antennas, except for practical considerations like the size and shape of the antenna. Circularly polarized antennas are usually much bigger than the linearly polarized ones.


Comments

Jtveg

Comment #1: Wed, 31 Aug 2016, 10:21 (GMT+1, DST)

Quote


Thanks for the information.

Koekebeest

Comment #2: Wed, 31 Aug 2016, 13:12 (GMT+1, DST)

Quote


I'm very curious about the antennas you send to Bruce. Very excited stuff!

Willem

Comment #3: Wed, 31 Aug 2016, 14:43 (GMT+1, DST)

Quote


Make that two of us, i'm also from Belgium, and always looking for more Fpv range :)

Quote: Koekebeest

I'm very curious about the antennas you send to Bruce. Very excited stuff!

Rocketnutz

Comment #4: Wed, 31 Aug 2016, 17:10 (GMT+1, DST)

Quote


Hello,
I'm very interested in your antenna design. Watched the mail drop from Bruce @ RCModelReviews, and looking forward to some real world testing.
I'm eager to test your antenna myself - please let me know if you are producing additional sample sets.
Thanks!

Dcara55135

Comment #5: Thu, 19 Jul 2018, 1:24 (GMT+1, DST)

Quote


Here is an interesting observation I do not understand. I have 2 RHCP patch antennas.

1) I point the 2 antennas at each other (stright in line boresights) with one as Tx and one as Rx (1600 MHz). I observe only freespace path loss of the correct value for the distance (5 ft). I actually expected some rejection due to crosspol (maybe 20 dB?) thinking if it received RHCP it would transmit LHCP.

2) Believing now that the observation implied a Rx RHCP antenna would also Tx RHCP I moved the Tx antenna next to the Rx antenna (2ft away) facing the same direction as the Rx antenna (parallel boresights) so now the transmitted wave would be rotating opposite of the previous test and I would see the cross pol and beam width rejection. Alas it was not the case. In short, with the Tx antenna in this position facing the same direction as the Rx was facing showed 10dB more signal at the Rx antenna than when I turned the Tx antenna 180 degrees facing the opposite direction.

Do you have an idea of what is happening?

Last modified: Thu, 26 Jul 2018, 19:00 (GMT+1, DST)

Maarten Baert

Administrator

Comment #6: Thu, 26 Jul 2018, 19:00 (GMT+1, DST)

Quote


Quote: Dcara55135

Here is an interesting observation I do not understand. I have 2 RHCP patch antennas.

1) I point the 2 antennas at each other (stright in line boresights) with one as Tx and one as Rx (1600 MHz). I observe only freespace path loss of the correct value for the distance (5 ft). I actually expected some rejection due to crosspol (maybe 20 dB?) thinking if it received RHCP it would transmit LHCP.

2) Believing now that the observation implied a Rx RHCP antenna would also Tx RHCP I moved the Tx antenna next to the Rx antenna (2ft away) facing the same direction as the Rx antenna (parallel boresights) so now the transmitted wave would be rotating opposite of the previous test and I would see the cross pol and beam width rejection. Alas it was not the case. In short, with the Tx antenna in this position facing the same direction as the Rx was facing showed 10dB more signal at the Rx antenna than when I turned the Tx antenna 180 degrees facing the opposite direction.

Do you have an idea of what is happening?

A RHCP antenna will always either produce or receive RHCP waves, regardless of whether you use it as a transmitter or a receiver. It's a bit counter-intuitive, but you can think of a receiving antenna as a time-reversed transmitting antenna: if you could make a video of the electromagnetic fields, a receiving antenna would look very similar to a transmitting antenna except the video is played backwards. So a clockwise rotation at the TX side becomes counter-clockwise on the RX side, which cancels out the fact that the antennas are facing in opposite directions (i.e. towards each other), so the direction of rotation ends up being the same on both sides. So in short, both the TX and RX side should use the same antenna polarization (LHCP or RHCP).

I don't fully understand the description of the second test - it sounds like the patch antennas are no longer facing each other. In that case you won't get much of a signal because you are outside the main beam of the antenna.

Last modified: Thu, 26 Jul 2018, 19:09 (GMT+1, DST)

Write a comment