Jump to content

Radio Blackout Detection?


Herbrax

Recommended Posts

Hello Everyone, 

So, in short, I'm Herbrax, kind of new to the forum, your classic space-nerd but do not have much of an academic background besides two classes on it, I'm a CS Major.
That brings me to my other topic, I'm on my Internship hunt and currently doing small projects to get a better understanding of Machine Learning Frameworks and make my resume more attractive.

Recently I've bought an RTL-SDR antenna to measure radio signal strength and signal to noise ratios ; my hope is to train a ML model to recognize patterns of radio blackouts to hopefully measure real-time solar storms.

I'm pretty sure there is already a dozen ways to do that but I wanted to give a shot to my approach for the sake of it. I do not have much of a scientific background hence I'm writing here in hope to get some clues and hints on how you would approach such a task, which frequencies would you focus on etc; I'm currently measuring and gathering data between 5 and 25Hz  

In hope of doing supervised training, I'd also need real-time data on the current radio blackouts, do you know where I'd be able to get such data?

Thanks !

Best regards,
Simo


TLDR : I'm making a ML model to recognize blackouts and would love your input :)

  • Like 1
Link to comment
Share on other sites

2 hours ago, Herbrax said:

Hello Everyone, 

So, in short, I'm Herbrax, kind of new to the forum, your classic space-nerd but do not have much of an academic background besides two classes on it, I'm a CS Major.
That brings me to my other topic, I'm on my Internship hunt and currently doing small projects to get a better understanding of Machine Learning Frameworks and make my resume more attractive.

Recently I've bought an RTL-SDR antenna to measure radio signal strength and signal to noise ratios ; my hope is to train a ML model to recognize patterns of radio blackouts to hopefully measure real-time solar storms.

I'm pretty sure there is already a dozen ways to do that but I wanted to give a shot to my approach for the sake of it. I do not have much of a scientific background hence I'm writing here in hope to get some clues and hints on how you would approach such a task, which frequencies would you focus on etc; I'm currently measuring and gathering data between 5 and 25Hz  

In hope of doing supervised training, I'd also need real-time data on the current radio blackouts, do you know where I'd be able to get such data?

Thanks !

Best regards,
Simo


TLDR : I'm making a ML model to recognize blackouts and would love your input :)

Sounds interesting. Typically these are immediately registered on D-RAP when hard X radiation hits our sunlit side.   Also our ionospheric D layer is affected during particle storms as well.  Typically for up to three days.  Largely over our polar regions.  NOAA handles this data in the USA.  These events should provide you with an easy way to cross check your data against theirs for accuracy and possibly calibration. Mike. 

Edited by hamateur 1953
Link to comment
Share on other sites

You could sample, record, and process WWV radio frequencies.  Starting 8 minutes after a strong solar flare, you may detect a blackout when the S/N or total S+N of the received WWV transmission decreases suddenly.

https://www.nist.gov/pml/time-and-frequency-division/time-distribution/radio-station-wwv

 

 

 

 

 

 

 

 

 

  • Like 1
Link to comment
Share on other sites

I am an amateur radio operator on the West coast in Northern California, and am sometimes unhappy with the various predictions done by others, using various models.  What I have found to be useful is to look at the realtime GEOS X-Ray Flux from NASA, and compare it to radio spotting on WSPRnet.org, which shows a global network of spotting beacons on various frequencies (the other charts and graphs on SpaceWeather Live are also useful).  I have several WSPR beacon transmitters, but you don't need a license to look at the WSPR map, which will show you the propagation beacons everywhere in realtime on the frequencies you are interested in.  I have seen that when the  X-Ray flux goes shooting up as a result of a solar flare, the propagation patterns will change very rapidly, whether it produces a blackout, or if the impact on Earth is smaller, the propagation will be enhanced.  For example, today, 7/11/24, there have been 5 M flares.  The result has been very poor propagation, with paths restricted to about 1500 kms (except for the path to Alaska-always good from here).  At the moment, looking at my beacon spots, there are 6 spots with 2 beacons transmitting on 14 MHz  and 28 MHz, with the average during calm space weather being about 25 spots at any particular time during the daytime.  Also, this time of day, around 00 hours UTC, there should be good propagation to the East Coast, but not today!  My call sign is W5OM.  If you want to see my beacons, you just enter the call sign where indicated on the WSPRnet page. If you try to look at all of the beacons, there are too many, and you won't see any patterns.  So select the call signs of stations in the paths you are interested in and on the bands you wish to look at, and look at their results.  This should help you determine what frequencies to measure with your RTL-SDR.

There are links to very good tools that should help you on the WSPRnet home page.  The M0XDK Map link on the left side of the page will give you realtime data on radio conditions and takes you to this page:  dx.qsl.net/propagation/propagation.html. That page has many different tools on it, which you may find useful.

I also do what Drax Spacex does and check WWV frequently.  If you can't hear WWV, that is indeed a radio blackout!

Good luck on your project.  If you find a model that works, there are thousands of people who would want to know about it!

  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you also agree to our Terms of Use and our Privacy Policy.