(SDR#) Baseband Recording up to 2GB ?

Status
Not open for further replies.

Your_account

Completely Banned for the Greater Good
Banned
Joined
Jul 13, 2013
Messages
1,259
Location
.AT
Hey
I want to record the Baseband over 1h oder longer white out any Gap between.
The problem is SDR# Stop at 2GB the are reach at ~2Min. The Disk Space is no Problem i have up to 12TB HDD Space. So how i can i Record it?
Thanks!
 

br0adband

Member
Joined
Apr 8, 2005
Messages
1,569
Location
Springfield MO
He wants to record the baseband audio using SDR# meaning the entire spectrum (for an RTL-based stick, that can be up to 2.4 MHz wide without dropping samples) and it consumes storage space incredibly fast. Apparently SDR# has some limitation of 2GB max for recording baseband audio...

If you must record the baseband audio then I'd suggest using SDR-Radio as it has no limitations in that respect and can even break the recordings into segments at various sizes. SDR-Radio also offers a decimator which you can use to go back to those baseband recordings and extract the audio content from a specific frequency which is pretty damned cool to be able to do. As the docs show, recording just 1 MHz of baseband for 48 hours would end up being ~346GB in size which is considerable, but using the decimator to pull content from that baseband for a single frequency could result in a file about 5-6GB in size (at 16 kHz sampling rate).

Also, this shouldn't be in this subforum - I know it was moved - it's more accurately an issue with an SDR program directly, has nothing to do with digital speech or voice decoding, should be in the Software Defined Radio subforum more than likely.
 

Pyr8

Member
Joined
Jan 30, 2006
Messages
29
Location
Mandalore system :|]
He wants to record the baseband audio using SDR# meaning the entire spectrum (for an RTL-based stick, that can be up to 2.4 MHz wide without dropping samples) and it consumes storage space incredibly fast. Apparently SDR# has some limitation of 2GB max for recording baseband audio...

If you must record the baseband audio then I'd suggest using SDR-Radio as it has no limitations in that respect and can even break the recordings into segments at various sizes. SDR-Radio also offers a decimator which you can use to go back to those baseband recordings and extract the audio content from a specific frequency which is pretty damned cool to be able to do. As the docs show, recording just 1 MHz of baseband for 48 hours would end up being ~346GB in size which is considerable, but using the decimator to pull content from that baseband for a single frequency could result in a file about 5-6GB in size (at 16 kHz sampling rate).

Also, this shouldn't be in this subforum - I know it was moved - it's more accurately an issue with an SDR program directly, has nothing to do with digital speech or voice decoding, should be in the Software Defined Radio subforum more than likely.
DAMN!!!!

That's great for every 3 letter agency known to exist. They can record the entire spectrum, and go back in time as far back as they have space on a server farm and pull up old com's. Good to know! ;)
 

Your_account

Completely Banned for the Greater Good
Banned
Joined
Jul 13, 2013
Messages
1,259
Location
.AT
i think a 4TB HDD Cost in Germany ~100€ at this time so you get for 400€ a nice Raid...
 

Token

Member
Premium Subscriber
Joined
Jun 18, 2010
Messages
2,070
Location
Mojave Desert, California, USA
DAMN!!!!

That's great for every 3 letter agency known to exist. They can record the entire spectrum, and go back in time as far back as they have space on a server farm and pull up old com's. Good to know! ;)
Why just three letter agencies? I pretty regularly recorded the entire 2000 to 30000 kHz spectrum at various times when searching for a station I know is on but that I can't find the freq for. With my current SDR that means I must make 7 recordings to grab that, it only does a maximum of 4 MHz width at a time. However, there is a new one on the market that will do 32 MHz at a time. True, it eats up some hard drive space, but hard drives are cheap ;).

Now, if I could just automate the search ...

T!
 

Pyr8

Member
Joined
Jan 30, 2006
Messages
29
Location
Mandalore system :|]
Why just three letter agencies? I pretty regularly recorded the entire 2000 to 30000 kHz spectrum at various times when searching for a station I know is on but that I can't find the freq for. With my current SDR that means I must make 7 recordings to grab that, it only does a maximum of 4 MHz width at a time. However, there is a new one on the market that will do 32 MHz at a time. True, it eats up some hard drive space, but hard drives are cheap ;).

Now, if I could just automate the search ...

T!
Use some voice/word recognition software to pull up conversations based on keywords! ;)
 

Your_account

Completely Banned for the Greater Good
Banned
Joined
Jul 13, 2013
Messages
1,259
Location
.AT
Use some voice/word recognition software to pull up conversations based on keywords!
lol for me as an Austrian its hard to unserstand many of the Austrian Slang. 99% of the voice recognition software understand only Germans...
 

Token

Member
Premium Subscriber
Joined
Jun 18, 2010
Messages
2,070
Location
Mojave Desert, California, USA
Use some voice/word recognition software to pull up conversations based on keywords! ;)

Not even close to that simple. Remember, we are talking files that are recording entire chunks of spectrum directly to the hard drive. This is not a simple audio stream of detected sound.

To use voice/word recognition to cue frequencies and transmissions of interest first the software would have to look at every frequency in the recorded range, with some specific minimum granularity. It then would look at every signal in that range. It must be able to identify real transmissions from noise. It would analyze each real signal and determine mode of transmission (USB, LSB, DSB, AM, CW, Pulsed, etc). It would then determine the type of transmission (voice, data, what type of data). And after all of that it would then try to automatically identify features, such as words, to key on.

I think something a bit more complex and fused than simple voice/word recognition would be in order ;) Of course if the task can be adequately defined it can then be done.

T!
 
Status
Not open for further replies.
Top