So, I have been turning my attention to the idea of buffering 24 hour grabs of all radio spectrum, using Web SDR. I'm just reaching out in case people are sitting on sets of these from times past. DM or [email protected].

Am I really to the point of archiving radiation? Yep.

The big concern is disk space. Under current approaches, a full spectrum grab of 24 hours of signal in one geographical location is ten terabyes! But one day a year might be worth it. For history.

If people don't know, we already grab 11,000 hours of radio a day worldwide, but that's not the same as radio spectrum.
@textfiles ohh, where do I find that/is there a way to search by station-year? I've been (passively) trying to track down a particular rendition of Here Comes The Sun that was used frequently as a bumper on WNPR around 2007-2010.
@textfiles
Is this all under a specific collection?
@textfiles a logical extension to internet archives mission. And really 10T / day is peanuts for such unique data. Maybe get some grant for it.
@textfiles I love the idea. Maybe recording samples of one hour every month and storing the complete speech2text would be more useful (especially allowing indexing and searching).
@a I've begun discussing things with people who know better to see if compression is possible.
@textfiles @a it would be pretty hilarious if it ends up that you can just use, say, FLAC and get ok compression.

@kepstin @textfiles @a You actually can use FLAC to some extent. I highly recommend reaching out to the people working with ld-decode and vhs-decode, as that's their primary domain of expertise; RF capture + post-capture compression

https://github.com/happycube/ld-decode
https://github.com/oyvindln/vhs-decode

They have a discord here: https://discord.gg/pVVrrxd

Or join #domesday86 on irc://irc.libera.chat

#preservation #rf #domesday86 #DomesdayDuplicator #archival #flac #datahoarder

GitHub - happycube/ld-decode: Software defined LaserDisc decoder

Software defined LaserDisc decoder. Contribute to happycube/ld-decode development by creating an account on GitHub.

GitHub
@textfiles image description: visualization of radio patterns with options to modify frequency, bandwidth, and other settings.
@textfiles That sounds high. Can you do an adaptive sample rate thing? You don't really need 8Gsps or whatever for LF.
@textfiles You could also save the 10TB annually for full IQ decoding, but do waterfall plots all the time to show traffic.
@textfiles what frequency range and sample rate are you capturing?

@textfiles How relevant is whitespace? Maybe some automated post process to clip whitespace?

Source -> ML(ID/Clip space) -> disk?

If you could specify KPI of the wave, record only those and specify everything else as non relevant or non eventful. Maybe add a buffer of the head and tail for each event.

@polygonpusher i have not gotten that far, I'm a stranger in these parts, recognizing some potential
@textfiles Understanding your signal to noise I think would be a great starting point. That will at least help you understand what can be ignored and what is of interest. With that you can optimize.
@textfiles What opportunities are there, if any, to compress this data?
@textfiles
I would imagine that there is a significant amount of white noise in it, isn't it?
@textfiles this kind of project could have some astronomical applications too if you can pick up the right frequencies.
@ch00f oh, there's a thousand uses
@textfiles It might be a good idea to use the #WebSDR tag. Not sure if I like your idea tho. You you know if the NSA is doing this?
@textfiles What format would this be stored in? Sounds like an interesting signal processing challenge to find a still useful but more compressed format
HOPE XII: Time Travel With Software Defined Radio

It’s easy to dismiss radio as little more than background noise while we drive.  At worst you might even think it’s just another method for advertisers to peddle their wares. But in rea…

Hackaday
@textfiles That sounds awesome, but be careful. Some parts are illegal to record in the US, namely cell phone bands, because it is considered wiretapping. Militaries would probably have something to say about it too.
@textfiles are you looking for cheap DVB-T TV dongles?

@textfiles Would be fun if we could have some special events... but... having that feature will make it hard to not save more... and more!!
I like the idea!!
I would like to have a day when there are some contests, i.e CQ WW cw/ssb https://www.cqww.com/
Maybe a day when the Grimeton transmitter is active
I would have loved to see the spectrum on 24/2 2022.. Or the days before/after

Hmm.. what do you mean by spectrum btw??

CQ World Wide DX Contest - Home

@textfiles It sounds like a great idea, but as you say, storage is massive. I'm assuming that you can't really reduce that size without losing data?
@JigmeDatse I appear to have woken up some pretty smart people so we are gonna see what we can do.
@textfiles It might be worth looking into ZFS file system. It offers high levels of compression for certain file types, and can also do deduplication. But I’m no expert. Only enthusiast. I highly recommend https://arstechnica.com/information-technology/2020/05/zfs-101-understanding-zfs-storage-and-performance/ as a starter point for anyone interested.
ZFS 101—Understanding ZFS storage and performance

Learn to get the most out of your ZFS filesystem in our new series on storage fundamentals.

Ars Technica
@textfiles Ok, I'm intrigued, tell me more. Specifically: what, and why. But also other things. I want to know.

@textfiles That is a very cool idea!

On the plus side, 10TB stored on AWS Glacier deep archive is about $10/month, so a day a year is doable for sure.

@textfiles Jason, you may have a look at https://github.com/oe2lsp/SDR-Buffer and check out the SDR Buffer instance located in Graz, Styria at the technical university: https://websdr.iks.tugraz.at/sdrbuffer/
GitHub - oe2lsp/SDR-Buffer: spectrum buffering over time, visualisation and playback

spectrum buffering over time, visualisation and playback - GitHub - oe2lsp/SDR-Buffer: spectrum buffering over time, visualisation and playback

GitHub
The Circle of HOPE (2018): Creating a Radio Time Machine: Software Defined Radios

YouTube