How do antennas work

[DrRobin]

Something else that helps, if you connect your phone/pad to your controller via a cable, then turn off WiFi/Bluetooth on your phone/tablet, these can cause a fair bit of interference especially in 2.45GHz.

You might also want to put the phone in flight mode to kill the cellular stages as well, the cellular modem in your phone might cause interference.

If you connect to your controller via WiFi, turn off the cellular and Bluetooth.

These measures will reduce the amount of interference generated by your phone in very close proximity to the controller and should reduce noise and therefore improve the video feed.

 

[WithTheBirds]

If you double the distance from 100ft to 200ft then the interferring signal will go down by 6dB. So if there aren't any other noise sources in the area then yes your signal to noise will improve. The chances of there been no other interference except in one spot is unlikely.

However, if you are 1,000m away your free space path loss is 100dB. If the interference souce is only 100ft away it's free space path loss is 70dB.

Assume that the controller signal and interference source are similar powers and antennas, increasing the interference source path loss from 70dB to 76dB isn't going to make much difference, noise is still 24dB higher than signal.

There is usually more noise in the 2.45GHz band, especially near built up areas, so if you are in an area with no trees, like a desert or over water, 5GHz might perform better if there is noise.

Note I am not suggesting there might not be other signal sources, I am saying there wont be a significant number that might become visible within relatively close proximity.

Here is the thing- increasing height does increase connection quality for a given range in RF congested environments. I was hoping you might assist in clarifying why.

 

[DrRobin]

Microwave signals are assumed to be line of sight, however, the signal spreads from the optical line of sight, it is called the Fresnel Zone. Objects in the Fresnel Zone cause attenuation, see the following link

Fresnel zone - Wikipedia

en.m.wikipedia.org

The size of the Fresnel Zone is greatest in the middle of the path and it’s size increases with the length of the signal path and reduces with frequency.. You could Google for a online Fresnel Zone calculator (everything RF) but at 2.45GHz and signal path of 1,000m it is going to be quite large in the middle, about 11m across.

So if your signal path is close to the ground, then the ground will be in the Fresnel Zone and cause attenuation, so increasing height will move the Fresnel Zone in to the clear and reduce attenuation.

The other problem would be multipath, a direct signal and a reflection, these would combine at the drone and give fading or reduction in signal strength. If the reflections were changing quickly, e.g. from cars on a road the received signal would vary and this might cause amplitude and phase distortion which would lead to errors in the decoded signal and a loss of quality. Increasing height would reduce the likelihood of multipath.

Just going back to your heights, at 100ft the visible horizon will be 12.3 miles in every direction, that is 476 square miles.

Increase the height to 200ft, the visible horizon goes up to 17.3 miles or 940 square miles, a lot of area.

So if you are out in the sticks away from WiFi, microwave ovens and other ‘in band’ noise sources, say 12 miles away from anything, then yes increasing height should improve the signal.

If however, you are near a built up area, the extra height is likely to work against you. Moral of this story fly out in the desert or over water and at least you don’t conflict with other drone regulations.

 

[DrRobin]

Also if extra height improves your signal path, it also improves the signal path of interfering sources.

 

[WithTheBirds]

Also if extra height improves your signal path, it also improves the signal path of interfering sources.

A typical wifi router at 17 miles away would be so far down in level (30 times as far away as the linked TX) I can't see it being an issue of significance. I take it you are factoring in point to point links, radar etc. I am not missing the point concerning increasing the height of the RX bringing more RF sources into view- I am inclined to think your fresnel zone explanation might prove to be of greatest relevance. The reason I had discounted it in my own earlier considerations was that when flying in the desert increasing altitude seemed to provide no benefit (could easily fly to the limit of available battery power without any connection issues).

Does the fact the MIMO is implemented reduce the effects attributed to the multi path consideration?

Edit- I meant to add, thank you for your insights here.

 

[DrRobin]

MIMO is Multiple In, Multiple Out. True MIMO has multiple transmitters and receivers although usually only has one of each, but with seperate modulation and initial decode stages, one for each antenna.

MIMO is used to transmit and receive to multiple devices at the same time, or by the use of multiple polarisations to increase the bandwidth by sending multiple streams to the same device. In fact to use the fast speeds on N/AC you have to have MIMO capability.

MIMO wouldn't actually do very much against noise and multipath but might improve throughput when you have a [relatively] good signal. The exception to this would be two antennas in different polarisations, one vertical and one horizontal as noise tends not to be in one polarisation, so you might get a bit of noise cancelling using MIMO.

Whilst the controller has two antennas, I don't know how many antennas the Mavic has, it might be different on different models, so if it only has one, you might not actually get any benefit from MIMO even if it uses it.

Multipath reflections often get their polarisation corrupted when they reflect, so it might start out as vertical polarisation, but end up as horizontal when it gets to the drone. Using MIMO and two polarisations might help here.

Rather than MIMO, I suspect that the two antennas on the controller and on the Mavic (if it has them) are for Noise Cancellation. Basically the wanted signal will be received mostly in one antenna, but the noise will be received in both. It is then fairly easy to cancel some of the noise. iPhones use this as I am sure that the majority of other smart phones as well as you home router also employ this technique. It is the only way to get the 'headline' speeds.

I used to angle my controller antennas at about 30 degrees from the vertical, one to each side, just to improve the signal in every direction and polarisation, although how much difference it makes I don't know. Perhaps someone could test with both antennas vertical, at 30 degrees and one vertical/one horizontal. The Mavic should be a little way away (250m) and be at 90 degrees to the controller antennas (off the side), then look at signal strength or video quality.

 

[Dr. Bill 1969]

up to a point VHF navigation works that way but only up to a point. It is line of sight. but there is a lot of traffic on it. It would seem to me WIFI would be the same. Here in Southern California the control tower KWJF had to change tower freqs because of interference from high altitude jet traffic in Mexico.Jet Traffic talking to Mazatlan center..as you know, there are 720 channels or freqs and for comm. and less for Nav..they are getting a little crowded.

The subject of radio frequency propagation is very involved and requires, at the minimum, more than a cursory study. Once you've got a basic understanding of RF propagation a study of antennas is in order.

DJI drones don't operate in the VHF band. The VHF band is from 30 MHZ to 300 MHZ. In the VHF band, electromagnetic fields are affected by the earth's ionosphere and troposphere.

DJI drones operate in the UHF frequency spectrum which is much larger than the VHF band. UHF waves are more affected by attenuation than VHF waves. Therefore, VHF waves can travel longer distances than UHF.

For anyone wishing to educate themselves in RF propagation:Wave propagation : Ground or surface wave, Sky wave, Space wave

 

[DrRobin]

The subject of radio frequency propagation is very involved and requires, at the minimum, more than a cursory study. Once you've got a basic understanding of RF propagation a study of antennas is in order.

DJI drones don't operate in the VHF band. The VHF band is from 30 MHZ to 300 MHZ. In the VHF band, electromagnetic fields are affected by the earth's ionosphere and troposphere.

DJI drones operate in the UHF frequency spectrum which is much larger than the VHF band. UHF waves are more affected by attenuation than VHF waves. Therefore, VHF waves can travel longer distances than UHF.

For anyone wishing to educate themselves in RF propagation:Wave propagation : Ground or surface wave, Sky wave, Space wave

VHF and UHF long range comms are reliant on either Sporadic E (exists at VHF, but hardly ever at UHF and certainly not at 2.45GHz) or Tropospheric Scatter, again this can extend in to UHF but it would have to be a major event for it to get to 2.45GHz, or you would have to be running very high power (kW). There is meteor scatter, but that is very short lived bursts at best, again probably difficult for 2.45GHz.

So we can consider 2.45GHz as line of sight and rule out most reflections, unless you have a large metal building or something similar.

As an aside I have an ADS-B Rx (1090MHz), I get signals from aircraft up to 280nm away every day, but at 38,000 ft, that is my horizon, again line of sight comms.

 

[Bernie707]

I would think if the mini is overhead at a height of 400 feet or less the need to have the antennas horizontal shouldn’t make much difference unless you’re under a tree some other obstacle, right?

 

[Thomas B]

I would think if the mini is overhead at a height of 400 feet or less the need to have the antennas horizontal shouldn’t make much difference unless you’re under a tree some other obstacle, right?

The controller signal from the antennae is broadcast from the flat sides. I do a lot of sunset vids where I go straight up from a TO position about 10-12 ‘ from the controller. If I don’t put the antennae in a horizontal orientation with flat long side toward the drone, or hold the controller tilted back to get this orientation I invariably get a warning popup to change the antenna orientation. Check you manual, it’s in there too.

 

[Camino Ken]

Hi @DrRobin

As a retired electrical (electronics) engineer with all of my experience in semiconductors I have an intellectual and practical interest in the Mavic RF communications. I am interested in your thoughts on a few things due to your in-depth and extensive antenna theory knowledge. The Mavics use what DJI calls Occusync technology for communications between the controller and the drone. In the Mavic 2 series it is Occusync 2.0. From what I have found it uses OFDM modulation and frequency hopping to increase the robustness of the link. In my opinion these should significantly improve resistance to interference. My question(s) are; do you think the dual antennas in both the controller and the drone are configured as an array with two elements to improve the directionality, and therefore distance, or more as a redundancy for diversity? By the way, from what I have read the antenna in the drone are located in the two front legs of the aircraft.

What do you think?

 

[douglas_in_philly]

Great video

Cheers
Alexander
Munich

Thanks for posting this. It was very informative!

 

[B52-D]

Just to explode some myths:
- VHF is 30 - 300MHz, UHF is 300 - 3000MHz.
- Mavic's run at 2450 or 5100MHz, the operate in the WiFi bands, so whilst one band is technically UHF it is usually refered to microwave or WiFi now.

- When you gain a lot of height you get a really good signal path to lots of noise sources (WiFi, baby monitors, CCTV, microwave ovens, other drones and video senders), the increased distance doesn't over-come the really good signal path, so generally you get more interference with height.

- Power lines and other HF transmitters (< 30MHz) do not cause any interference at 2.45 or 5GHz no matter how powerful they are, if they did they would be operating outside their licence terms. The do interfere with the compass though.

- Some radars can cause interference, especially at 5GHz, these are typically rainfall radar and operate near airports, another reason to avoid these areas.

- There is a lot less interference in rural areas.

- 2.45GHz has less loss from building ands trees/bushes. 5GHz suffers very badly from trees, but does reflect better from buildings. So if you want long range, use 2.45GHz, if you get the choice.

- Antennas have a polarisation, if the you are crossed-polarised you will get a lot of attenuation. The stub antennas on the controller are a 1/4 wave stub at 2.45GHz and possibly a 3/8 wave at 5GHz. When they are vertical, you get all round coverage (best straight off the side), but little straight off the top, that is why it is worth tilting them when the drone is over-head. I am not sure what the drone antenna is, I would assme some sort of patch antenna with hemispherical coverage. It might be circular polarised to give all round coverage regardless of orientation.

- Antenna reflectors do work if they are designed correctly.

- Using the controller antennas at 45 degrees gives you vertical and horizontal polarisation and better all round coverage at the expense of a slight bit of gain, perhaps only 3 or 4dB.

- To get best range, make sure you choose the best channel with least interference (if available on your model), use 2.45GHz (if available), orientate the controller and antennas to ensure that the drone is straight off the side of the antennas (even more important if using reflectors) and make sure they are vertical.

How do I know all of this, my higher degree was in antenna design at 2.45GHz and I design antennas at these frequencies and deploy comms systems in these bands for work, I have done this for over 35 years, nothing much changes in the basic science.

Amen.