Can DSD (or any other software) be used to pull data from digital voice formats?

[kb1ipd]

I am not sure if this is in the right sub-forum, and forgive me if it is not...

Nearly all digital voice formats also support data transmission. Sometimes, it seems to be added almost as an afterthought, and it seems that many don't utilize this feature.

For example, P25 is generally considered a voice format, but it can also be used for data. Early on, it seems that it was seen as being a unified solution for voice and data to mobile data terminals etc. Given that it's pretty slow in this regard, this feature seems less of a selling point in recent years. Similarly, NXDN can have data channels as it does voice. Provoice, dPMR have these options.

A few formats seem to put the data capability more front and center. Opensky is certainly oriented toward both voice and data communications and so is D-Star. D-Star apparently can operate at the comparatively high data rate of 128 kbps, at least on some bands.

Some implementations of these voice modes may be able to pair multiple channels for greater capacity when used for data transmission.


From what I can see, the data and voice modes are transmitted via the same basic modulation and encoding, except it's not voice packets riding the bit stream in data mode.

Is DSD (or for that matter, any other software package) "smart" enough to know when it has hit upon a data-mode transmission stream? And if so, is it possible to decode any of the data?

If not, would it be hard to modify DSD or perhaps pipe the audio to something else?


(again, sorry if this is off topic for the voice forum, since it spans a couple areas)

 

[br0adband]

If I had to guess, I would say this: the reason DSD and DSD+ don't bother with decoding the data streams is that they could be encrypted to some degree, more so (and more often) than traditional voice content is or happens to be. It's obvious that any transmission of digital data contains all of it, meaning the data for use with MDTs, or simple text messages sent from unit to unit, etc, as well as the traditional voice comm data too in either openly decodable format or encrypted. My belief is that the data transmissions are always encrypted but I have no absolute proof of that.

As far as you wondering if DSD or DSD+ can "know" or if they're smart enough to understand the difference, of course they are, but the primary focus for monitoring the comms is so the voice comms can be heard, I suppose. Perfect example: DMRDecode, which isn't used itself for direct decoding of the data or voice content but easily notes it when it exists by showing us what's going on per slot/channel during a transmission in progress meaning - for example - if both slots are active it could show a Group call on Slot 0/Channel 1 while a Data call on Slot 1/Channel 2 is in progress (along with the source unit and target unit identifiers as well).

While I don't have any direct evidence, I still believe that data being transmitted in the digital bitstreams is more than likely encrypted, whereas voice content can be but (luckily) isn't; if some "privacy" mode is enabled on a given digital communications system then the voice content does get encrypted as well.

Conjecture on my part, of course, but I'm sure someone with vastly more in-depth knowledge of these digital protocols will comment at some point with some of that knowledge (at least hopefully).

 

[racingfan360]

DMRDecode decodes DMR text messaging which is in the clear (but rarely used AFAIK)

 

[kb1ipd]

If I had to guess, I would say this: the reason DSD and DSD+ don't bother with decoding the data streams is that they could be encrypted to some degree, more so (and more often) than traditional voice content is or happens to be. It's obvious that any transmission of digital data contains all of it, meaning the data for use with MDTs, or simple text messages sent from unit to unit, etc, as well as the traditional voice comm data too in either openly decodable format or encrypted. My belief is that the data transmissions are always encrypted but I have no absolute proof of that.

I was wondering about that. It's hard to know, but I would not be surprised if it is not encrypted. You might be surprised to see how many transmissions are not encrypted because nobody seems to bother doing so. I guess it's "Security by obscurity" - why bother if so few people have the proper equipment to monitor anyway? Voice is usually unencrypted. And, of course, it would always be for something like ham radio.

I guess one of my other questions would be whether or not this feature of digital voice systems is even used very often. Many police departments have P25 radios, but most rely on 3G/4G cards for their mobile data (at least from what I have seen)

Is data on these radios one of those features that just never, or almost never, gets used by anyone? Or is it a common thing to make use of the data capabilities?