Mie scattering in clouds at 940 nm

[W1VLF]

TL;DR Summary

Looking for help with understanding MIE scattering in clouds and how it affects my communications project

How did you find PF?: Google search

[Mentor Note: Two threads merged]

I am an amateur radio operator call sign W1 VLF.

Currently I have been experimenting with optical communications using 940 NM light through free space. Having accomplished twenty-five mile communications links using infrared light, I am now interested in reflecting that same invisible 940NM light using clouds as a reflector, or scatterer to communicate between two locations that are not line of sight between each other.

I have questions about MIE scattering and can not find answers. My hope is to find someone on the forum that is versed in light scattering and optics to help me understand the the MIE scattering mechanism into optimize my communications links.

This is not a for profit venture this is just hobbyist level experiments and in search of knowledge on the topic.
Thank you,
Paul CIANCIOLO Amateur radio operator W1 VLF

 

[berkeman]

W1VLF, 73. KI6EGL

 

[W1VLF]

Hello all,
I am brand new to this forum and hopefully my questions are not too basic as I don't have a college degree in physics or any college degree for that matter. Let me explain what I have been doing experimentally and perhaps someone can offer some advice.

The experiment encompasses a communications link for sending Morse code modulated onto a 940 nanometer light source. This light source was received at a remote location using a lens and optical to electrical conversion over distances greater than 25 miles Over line of sight paths.

Recently experiments have taken me to using clouds as a reflective surface that the transmit and receive sites can both see even though a terrestrial path across land is obstructed. I understand from research that light scattering happens in clouds almost independent of wavelength causing the clouds to be white. Let me detail the experiment a little further

At location a is an infrared source pointing straight up into the sky perpendicular to the cloud and the Earth's surface. The receiving station is located at various distances away from the IR source, and the receiving station points to the Clouds directly above the transmitting source. The angle at which the receiving station must point, meaning the elevation angle is dependent upon the distance from the transmitting site, and the height of the clouds. Currently using this system I have been sending Morse code messages at a distance of four and a half miles from the transmit site using the clouds as a reflector or scatterer. There is a lot of signal strength above what's needed to make the communications possible so the distance of four and a half miles could be much greater until the angle of the receive station becomes so low that it cannot see over the local horizon.

So my questions pertain to how the light is scattering at the base of the cloud and even as it penetrates into the cloud and whether both transmit and receive sites would have less attenuation if pointed to a common point in the sky that was midway between the transmit and receive sites.

It's kind of difficult for me to verbalize the exact situation of my experiments, but the goal is to find if there are more optimal angles besides pointing the transmitter straight up and having to receive system point at a common point in the sky where both the vertical infrared source, transmitter and receiver elevation have a common volume. Using trigonometry I have been successful in finding the approximate angle to point the receiver at different distances from the transmitter and the math has actually worked out.

So if anyone has any insights into this or can help me visualize what's happening in the clouds I would greatly appreciate it.

Thank you very much for reading this

Paul Cianciolo W1VLF

 

[W1VLF]

Hello Mike!! Thanks for replying. Nice to see fellow ham on here.. I am a little bit intimidated by the level of questions asked here and worried I posted to the wrong are a or inappropriate question.

Thanks for saying hello it goes a long way to see another ham on here

Paul W1VLF

 

[W1VLF]

Yes,,,, Code is no longer a requirement, Sorry i missed that in your post. I worked at the ARRL for 3 years as an RFI engineer in the Lab, Been a ham since 1979

 

[berkeman]

I won't be much help on Mie scattering beyond linking to the Wikipedia article (which you probably have already read): https://en.wikipedia.org/wiki/Mie_scattering

But I do want ask if this is all being done at baseband modulation on the optical carrier. Are you just using On-Off keying to send the Morse code, or are you also sending a carrier wave to help with lowering the noise floor for the channel?

For example, your IR remote control for your TV set does not just send a train of pulses for each button push. Instead, the information is sent on a carrier wave (I forget the frequency), and at the receiver there is a narrowband filter at that carrier wave frequency to filter out noise at other frequencies. This is an important technique for lowering the amount of noise in the received signal, and it will improve the accuracy of your communication channel.

Further, the carrier frequency should be chosen to avoid natural noise sources in the channel. You will have a number of them in your atmospheric and Mie/cloud channel, so you should figure out what the noise characteristics are or do the experiments with a spectrum analyzer at the receiver in order to help you figure out the carrier frequency (or frequencies) that you should use.

Does that make sense? I can try to find some references in a few minutes if that would help. 73

 

[W1VLF]

Hi Mike,
I should have been a lot more specific about what's going on in the experiment. I'll try to answer those questions here. The LED emits 940 nanometer light, that light is switched on at a 50% duty cycle at a 14 kilohertz rate. 14 kilohertz is not critical it's just the frequency I had available at the time to switch the LEDs on and off.

To transmit the LED array at 940 nanometers it's turned on and off, as you say on off key to send Morse code to the fire end of the circuit. Typically a remote control for a television set for instance runs at 38 kilohertz or 40 kilohertz carrier frequency and is modulated with information that is specific to the appliance or tv that it is supposed to control.

I am doing all the things you've already mentioned for example using narrow bandpass optical filters, high impedance amplifiers as the transimpedance amplifier, even using avalanche photo diodes, because of their superior sensitivity to light. As I mentioned earlier I've been doing this for a while and now my concentration is on the actual medium, the clouds,, that are providing the reflecting, scattering medium.

What I don't understand about MIE scattering is how the angle of incident light on particles refracts or scatters, and how that would affect pointing angles from transmitter to receiver.

For instance if I point the transmitter straight U with a cloud height of around 2000 feet, at two miles away the elevation angle of the receiver is somewhere around 11 degrees. What I would like to find out is if I could aim the transmitter and receiver at a mutual cloud or group of clouds and how that energy would be scattered versus putting power onto the cloud at a perfect 90 degree angle it's kind of hard to explain I hope that helps should have been more descriptive in my initial query.

To your other point I do use a spectrum analyzer of sorts as the receiver. Actually an SDR, software defined receiver and it allows me to evaluate signal strength over time in various bandwidths Etcetera Etcetera
Thank you again

Paul W1VLF

 

[DaveE]

You're not planning on directing high power lasers into the sky are you? FAA and CDRH may be interested.

 

[berkeman]

You're not planning on directing high power lasers into the sky are you (FAA and CDRH may be interested)?

Good point. Do they care about 940nm IR lasers?

 

[W1VLF]

You're not planning on directing high power lasers into the sky are you? FAA and CDRH may be interested.

Of course not !! All my experiments are done with LEDs having a typical beamwidth 6 to 15 degrees. Not .057 degrees in a typical handheld laser. I am curious why you would ask me that question? Paul W1VLF

 

[DaveE]

Good point. Do they care about 940nm IR lasers?

Yes. They care about nearly all lasers. However, their requirements do depend on wavelength.
Anyway 940nm is well within the band they care most about. For example YAG at 1064nm is an extremely common high power laser.

 

[W1VLF]

Yes. They care about nearly all lasers. However, their requirements due depend on wavelength.
Anyway 940nm is well within the band they care most about. For example YAG at 1064nm is an extremely common high power laser.

I know I am new here, and folks don't know me but I never mentioned lasers high or low power at any wavelength

 

[DaveE]

I am curious why you would ask me that question?

Because that's the best way to do it. Seriously, you want a laser, you just probably can't have it. Like this air to submarine com project I worked on.

 

[DaveE]

These regulatory bodies also care about LEDs, searchlights, etc. But, as you say, the risk and requirements are much much lower because of the big divergence. In those cases it's the nearby risk that matters most.

 

[berkeman]

Good stuff.

14 kilohertz is not critical it's just the frequency I had available at the time to switch the LEDs on and off.

To your other point I do use a spectrum analyzer of sorts as the receiver. Actually an SDR, software defined receiver and it allows me to evaluate signal strength over time in various bandwidths

Have you done any experiments with a DC illumination source and the spectrum analyzer monitoring the received optical signal? What kind of noise frequencies did you see in the received optical signal? If you have a screenshot of that test, it would be interesting for me.

 

[W1VLF]

These regulatory bodies also care about LEDs, searchlights, etc. But, as you say, the risk and requirements are much much lower because of the big divergence. In those cases it's the nearby risk that matters most.

Laser pose too many problems for taste. Safety first, but after that they are not likely to be a good way to go to illuminate a cloud. At cloud height of 2000' a 1 milliradian divergence laser would illuminate a roughly 2' diameter spot on the clouds If the clouds were a flat homogenous surface this might be a reasonable way to go. I believe Lambertian scattering would apply here. But clouds are not so while a small spot may give good signals in a LOS situation or a flat reflective surface, not so with the clouds,

This is what I am trying to learn about. Another problem using a laser is that the field of view of my receivers is typically on the order of 1to 2 degree beamwidth this makes the field of view of the receiver about 80 ' at a cloud height of 2000' that's looking at a lot of potential noise in the form of background light. I would rather sacrifice the greater spot intensity provided by a laser, for the much larger coverage area on the cloud.

Even using 15 degree beamwidth on the transmitter the signal one can see a the receiver with is 2 degree beamwidth, there is lots of fading taking place as the cloud height varies from moment to moment. It is actually very interesting to watch.

Paul W1VLF

 

[W1VLF]

Noise from DC illumination shows up as noise, white noise or a decrease in Signal to noise ratio.

However back in the early days of these experiments when using baseband for instance listening to 0 to say 10 kilohertz there was lots of weird stuff reflected back from the clouds or even looking at the horizon stuff like sodium lights or any gas type of light will have have tremendous amount of harmonics as the gas switches on and off during the 60 Hertz sine wave. Actually what you'll see is 120 Hertz 240 Hertz 360 Hertz 480 Hertz et cetera et cetera et cetera. This is one of the primary reasons that I opted for a much higher frequency, 14 kilohertz. I have done some daytime testing but the the sun is such an incredible noise source.

There are some things that can be done to mitigate some of that for instance I had one link running with a single remote control type LED across 4 miles during the day by playing around with bias on the photo detector. But no amount of filtering or detector bias or anything allows you to compete with the sun. But there are a lot of odd things that you hear if you use baseband. Led display on a clock tv microwave whatever using a scanning or strobing scenario and they can be heard. Flashing tower lights from many miles away can be heard, even the smooth 60 Hertz Hum of red tower lights as they brighten and dim can be heard quite nicely.

That's the primary reason that I started using much higher frequencies there are no interfering noises. Another note folks who are into astronomy will understand DC noise as background noise or lack of contrast in the images that they want to see. Same goes for photographers if they're doing long exposures of the sky and the sky has any background noise it shows up as a less contrasted photograph

Paulc W1VLF

 

[berkeman]

It sounds like you are familiar with the channel noise avoidance issues I was thinking about.

For instance if I point the transmitter straight U with a cloud height of around 2000 feet, at two miles away the elevation angle of the receiver is somewhere around 11 degrees. What I would like to find out is if I could aim the transmitter and receiver at a mutual cloud or group of clouds and how that energy would be scattered versus putting power onto the cloud at a perfect 90 degree angle it's kind of hard to explain I hope that helps should have been more descriptive in my initial query.

I wonder if you could use a scanning mirror setup to sweep angles from vertical down toward the receiver to hit the best reflection angles. There are issues with the carrier frequency and other stuff, but it might help you deal with the varying geometry of the underside of the cloud cover.

Does your Tx/Rx scenario allow for both parties knowing the location of the other station so the Tx party can sweep from vertical toward the location of the Rx party?

 

[berkeman]

Oh, and to add -- can both parties cooperate on the channel to figure out the best communication angle versus time? There are some very interesting approaches to dynamic channel communication[1]*.

[1] Adaptive Signal Processing by Widrow and Stearns. (c)1985 Prentice Hall

* This book from my bookshelf is near the edge of my ability to understand, and I still have much of it to go through. A project like yours would be what would motivate me to finish going through it.

 

[hutchphd]

There is a fundamental question here: what is the OP trying to optimize ? There is a lot of geometrical structure in the Mie scattering and allowing the geometry to "float" may be an issue. So what problem are we (he) attempting to solve? "Better" can have myriad interpretations (lower power? higher bandwidth? less interference?) The answer will depend upon the exact question.
Mie scattering from spheres is a well-studied question (as it should be!) .

 

[DaveE]

What kind of modulation are/will you be using?

 

[Andy Resnick]

Currently I have been experimenting with optical communications using 940 NM light through free space. Having accomplished twenty-five mile communications links using infrared light, I am now interested in reflecting that same invisible 940NM light using clouds as a reflector, or scatterer to communicate between two locations that are not line of sight between each other.

I have questions about MIE scattering and can not find answers. My hope is to find someone on the forum that is versed in light scattering and optics to help me understand the the MIE scattering mechanism into optimize my communications links.

If I understand what you are trying to do, I'm not sure it will work. It seems like it should work- I can see high-power flashlights reflected off of low-lying clouds for example, but that would be amplitude modulation and is not how you encode your signal.

Your basic challenge is that the cloud is not a single reflective object- each of the constituent water drops (or ice crystals!) will act as an independent scattering source throughout the entire illuminated volume of the cloud. In addition, there will likely be multiple scattering events- Mie scattering is a single-scattering model- further degrading the signal.

You may find some insight or ideas by checking out the literature on imaging/communicating through turbid media (think muddy water or fog). AFAIK, those systems use light polarization to maintain a coherent communications link.

 

[Gleb1964]

Isn't 940nm close to the water absorption peak?

 

[DaveE]

Isn't 940nm close to the water absorption peak?

Reflectance? Shorter wavelengths tend to be a bit better, but it's harder to get high power sources.
Anyway, google...
https://www.researchgate.net/figure...grain-sizes-In-the-visible-and_fig1_307830440

 

[Gleb1964]

Apart from cloud reflectance, there is a significant path within the atmosphere where water vapor efficiently absorbs signals at 940 nm.

 

[W1VLF]

Hello Hutch,

Sorry for the delay I was away all weekend.
First thank you so much for taking the time to answer.

I agree I need to frame the question much better. Please bear with me as I am not a scientist and never went to college, just self learning .

2 attached screen shots.

The first is the current configuration of the experiment. Right angle trig says with the parameters height of the cloud and distance between transmitter and receiver the elevation angle at the RX should be approx. 8 degrees. In practice this works, after many tries with the clouds at this and other heights the math proves out.

I don't know what kind of scattering this is but I suspect not Mie scattering but I don't know. From what I researched the particle size of moisture in white clouds is somewhere 10,000 nM and 15,000 nM. Mie scattering would be more of a forward scattering with this size particle?

Screenshot 2 shows the same distance between the transmitter and receiver. However this time they are pointed at the midpoint of the 2 miles path. With this configuration both transmitter and receiver are pointed at roughly 21 degrees. 21 degrees allows for access to the cloud from many more places IE low elevations have obstacles, trees buildings etc blocking the path.

So I guess I would ask these questions
1) what type of scattering is taking place in/on the cloud ? Or is is a combination of Mie, Rayleigh, Lambertian scattering?

2) Would there be an advantage or disadvantage in reflected signal strength to orientating the light beam to the mid path as shown in screenshot 2?

I am assuming for this question that the line of sight losses between transmitter and cloud and the receiver and cloud at not considered. I am only interested in the reflection, scattering at the cloud itself.

I cannot do this experiment empirically as things change from moment to moment with the received signal strength.


Thank you so much again for your time..

Paul
W1VLF

 

[W1VLF]

Apart from cloud reflectance, there is a significant path within the atmosphere where water vapor efficiently absorbs signals at 940 nm.
View attachment 361644

Hello Gleb.. Yes we have found that in terrestrial experiments. Lower dew point means better signal strength. Same with the light our eyes can see. We have a water tower 10 miles distant that we use as a visual aide
Paul W1VLF

 

[hutchphd]

Again.
Is signal strength what you wish to maximize? What if it is quite variable? What about dispersion? I think you need to specify exacty what you wish to optimize. Why are you interested in this ?

 

[W1VLF]

Hello Hutch,

As ham radio operators we are very used to using less than optimal signal strengths. Quite variable would be fine if I knew how to somewhat predict what was going to happen to the returned light .

Is signal strength what you wish to maximize?
Yes that would be a desired outcome.

What if it is quite variable?
This is not a problem we deal with probability of reception all the time.

I think you need to specify exactly what you wish to optimize.
I wish to be able to understand, for a given cloud height and distance between transmitter and receiver whether there would be an signal strength advantage to point the transmitter straight up or at some angle between the 2 locations.

I am interested in purely to extend my range from transmitter and receiver and to understand the physics behind what it happening in the clouds.

Thank you
Paul
W1VLF

 

[Paul Colby]

My masters thesis involved Mei scattering. We were interested in quantifying the effects of multiple scattering on measurements of binary phase transitions in fluid mixtures. People used laser light to measure these systems which could become quite turbid. Anyway, cloud particles range from 5 to 50 microns and I assume are usually water. Mie series codes exist for dielectric spheres. You might rummage around on,

Mie Codes

Some of these will produce differential cross sections to your hearts content.