Author Topic: 1200 baud data transfer over audio passband of cellphone. Is that possible?  (Read 5903 times)

0 Members and 1 Guest are viewing this topic.

Offline sv3oraTopic starter

  • Regular Contributor
  • *
  • Posts: 84
  • Country: gr
So through wired phone there is no compression?
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
So through wired phone there is no compression?
You are advised to search The Internet. Answer is all over the place.
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 10034
  • Country: gb
Now, I do not know if this is due to the mobile phone codecs or the landline phone codecs
Landline usually have 8ksps @8bit PCM, no codecs. Who uses landline today anyway  :-//
Simple 8 bit samples would have a rather limited dynamic range for good audio transmission. Landlines have used a simple pseudo-logarithmic compression codec, G.711, since the very beginning of digitising the network in the late 1950s. The variant used in most of the world is called A-law, and the variant used in North America and a few other places is called u-law. They compress linear samples of 13 to 14 bits at 8ksps to samples of 8 bits at 8ksps. This causes significant distortion, but everything designed for the telephone network since the late 1950s has been specifically designed to operate reliably through these codecs, both used alone and in tandem.

From the 1980s some operators introduced ADPCM compression for their long distance links, which can handle slow modems, but not the fastest ones. More recently, VoIP has brought some very low bit rate codecs into the mix, which, as with cellular systems, mean all attempts to use standard modems fail. ISDN was supposed to bring the wideband G.722 codec to the telephone network in the late 1980s, but wideband voice didn't catch on until very recently with cellular and VoIP.
 
The following users thanked this post: ogden

Offline Psi

  • Super Contributor
  • ***
  • Posts: 10385
  • Country: nz
i wonder if you could craft the signal so the audio compression algorithm couldn't compress it a huge amount.
And get extra bandwidth out of the system that way.

Probably not, but its an interesting thought experiment.
Greek letter 'Psi' (not Pounds per Square Inch)
 

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 9333
  • Country: fi
i wonder if you could craft the signal so the audio compression algorithm couldn't compress it a huge amount.
And get extra bandwidth out of the system that way.

No, as shown by the earlier replies, GSM compression is a fixed rate (of 13 kbit/s).

You'll get the most bandwidth out of the system by carefully designing and optimizing the modulation scheme for the GSM compression.
 

Offline sv3oraTopic starter

  • Regular Contributor
  • *
  • Posts: 84
  • Country: gr
i wonder if you could craft the signal so the audio compression algorithm couldn't compress it a huge amount.
And get extra bandwidth out of the system that way.

No, as shown by the earlier replies, GSM compression is a fixed rate (of 13 kbit/s).

You'll get the most bandwidth out of the system by carefully designing and optimizing the modulation scheme for the GSM compression.

The problem is not the bandwidth, but the compression that seems to be applied by the codecs, which do not let the supposed tones pass for more than a few seconds.
 

Offline sv3oraTopic starter

  • Regular Contributor
  • *
  • Posts: 84
  • Country: gr
Do you think that I could pass the 1200 baud 1200Hz/2200Hz through the wired telephone instead, or will I have the same problems as the GSM?
 

Offline jhpadjustable

  • Frequent Contributor
  • **
  • Posts: 295
  • Country: us
  • Salt 'n' pepper beard
Do you think that I could pass the 1200 baud 1200Hz/2200Hz through the wired telephone instead, or will I have the same problems as the GSM?
Worth a try. The fewer switches and long-haul links your call has to pass through, the better your odds.

The problem is not the bandwidth, but the compression that seems to be applied by the codecs, which do not let the supposed tones pass for more than a few seconds.
Bandwidth is sometimes informally used to refer to data rate.
"There are more things in heaven and earth, Arduino, than are dreamt of in your philosophy."
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8526
It's worth seeking out the official GSM specs (they are dense but free) on codecs and reading them carefully, and also playing around with a PC-based simulator, it will save all the hassle of manipulating actual phones (unless you like that.)
 

Offline _echo

  • Newbie
  • Posts: 2
  • Country: us
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #34 on: November 04, 2019, 05:26:33 am »
Hello,
Not exactly 1200bps, but if you're trying to get binary data through the audio channel of a cell phone (or any phone for that matter), you might try borrowing the portions of the FSK encoding used by TTYs/TDD. Although their use is declining, this baudot infrastructure is still active today.

The core encoding is half duplex FSK, traditionally it's 45.4 bauds/bps. There are a few faster encodings out there, it's possible that those were included.

Telecom vocoders usually have hooks to play nice with TTY, some require a menu option to enable, but it's certainly present in most cell phones.

Sorry, I don't have any additional details. I thought that it might cause some others to remember the encoding.
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #35 on: November 04, 2019, 09:45:26 am »
Do you think that I could pass the 1200 baud 1200Hz/2200Hz through the wired telephone instead, or will I have the same problems as the GSM?
Should work fine. AFAIK, the landline network has the ability to detect modems and change something in the codecs to make them work. Fax machines have relied on this for decades.
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #36 on: November 04, 2019, 10:04:29 am »
AFAIK, the landline network has the ability to detect modems and change something in the codecs to make them work. Fax machines have relied on this for decades.
Really? Citation needed. My knowledge about landline codec (G.711) say that they do not "detect modems". I talk about landlines of modem/fax era, T1/E1 circuits.
 

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • !
  • Posts: 2699
  • Country: tr
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #37 on: November 04, 2019, 10:09:41 am »
Last time I tried (maybe 10 years ago) I could not fax from my land line anymore. The FAXes are a thing of the 80s!
The further a society drifts from truth, the more it will hate those who speak it.
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #38 on: November 04, 2019, 10:48:09 am »
Last time I tried (maybe 10 years ago) I could not fax from my land line anymore. The FAXes are a thing of the 80s!
Yes, so true. Today (cheap) telcos happen to use low rate codecs in their (mostly SIP) long haul trunks. On mobile networks you can literally hear cheap telco operator.

BTW I noticed "landline" solution which is made out of cellular/LTE terminal box with landline phone attached. Telco did not even bother to add any kind of UPS function, so your "landline phone" works only when you have mains electricity.  :palm:
 
The following users thanked this post: tooki

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #39 on: November 04, 2019, 11:16:12 am »
AFAIK, the landline network has the ability to detect modems and change something in the codecs to make them work. Fax machines have relied on this for decades.
Really? Citation needed. My knowledge about landline codec (G.711) say that they do not "detect modems". I talk about landlines of modem/fax era, T1/E1 circuits.
Oh gosh, it’s just a random thing I read or heard somewhere ages ago.

I don’t know what the magic search terms are to find out accurately. (Every search result about codecs is talking about VOIP, making it hard to find the docs from before that.) The closest I got is this, which is already VOIP era: https://www.eetimes.com/document.asp?doc_id=1201937

Maybe that’s what I heard, but I thought something similar applied to the PSTN, especially for long distance, where heavier compression was used, but fax still worked.


Last time I tried (maybe 10 years ago) I could not fax from my land line anymore. The FAXes are a thing of the 80s!
You would think, right?? But apparently in some countries and/or industries, they’re still widely used, especially in jurisdictions where faxes are considered legal documents, but emails and other internet transmissions are not.

I haven’t had a fax-capable device (ignoring my vintage computers with modems) since the 90s when I was living with my parents, so even when I had a landline of my own (which also has been a long time), I never had a chance to try faxing on it.
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 10034
  • Country: gb
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #40 on: November 04, 2019, 06:20:24 pm »
Do you think that I could pass the 1200 baud 1200Hz/2200Hz through the wired telephone instead, or will I have the same problems as the GSM?
Should work fine. AFAIK, the landline network has the ability to detect modems and change something in the codecs to make them work. Fax machines have relied on this for decades.
Why do people make up rubbish like this?
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #41 on: November 04, 2019, 10:16:56 pm »
Do you think that I could pass the 1200 baud 1200Hz/2200Hz through the wired telephone instead, or will I have the same problems as the GSM?
Should work fine. AFAIK, the landline network has the ability to detect modems and change something in the codecs to make them work. Fax machines have relied on this for decades.
Why do people make up rubbish like this?
The possibility of inaccurate recollection (as explained above) is not the same as “making up rubbish”. (And absent documentation either way, I’m not certain that my memory is actually wrong.)

Edit: documentation found, thanks to ollopa remembering the specifics. So yeah, I was right. Definitely not "rubbish", and definitely not "made up". (Also, how much more could I have couched it in uncertainty? I made it clear that I wasn't 100% sure, prefacing it with "AFAIK" and only saying that "something" got changed.)
« Last Edit: November 05, 2019, 11:25:24 am by tooki »
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #42 on: November 04, 2019, 10:59:21 pm »
Why do people make up rubbish like this?
Nah. We are all guilty of "imprecisions" here, in this same discussion. No need to start flames.
 
The following users thanked this post: tooki

Offline ollopa

  • Regular Contributor
  • *
  • Posts: 125
  • Country: 00
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #43 on: November 05, 2019, 12:36:55 am »
It's not rubbish.  One of the early phases of dial-up negotiation is signaling to the channel (AKA phone company) to disable echo cancellation for full-duplex operation.  That's completely in-line with "the landline network has the ability to detect modems and change something in the codecs to make them work."
 
The following users thanked this post: amyk, tooki, newbrain, I wanted a rude username

Offline westfw

  • Super Contributor
  • ***
  • Posts: 4349
  • Country: us
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #44 on: November 05, 2019, 11:09:00 am »
Iirc, the old 1200bps modems required some sort of phase shift modulation, rather than just fsk.
(For bi-directional, anyway.  The older 1200/150bps modems may have used straight fsk.)

 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #45 on: November 05, 2019, 11:18:19 am »
It's not rubbish.  One of the early phases of dial-up negotiation is signaling to the channel (AKA phone company) to disable echo cancellation for full-duplex operation.  That's completely in-line with "the landline network has the ability to detect modems and change something in the codecs to make them work."
Thank you! That's what it was.

Knowing the specifics, I was able to find more, e.g. ITU-T Recommendation G.16: Interaction aspects of signal processing network equipment, sections 5.2.1–5.2.2, pp 5–9.
« Last Edit: November 05, 2019, 11:23:31 am by tooki »
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 10034
  • Country: gb
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #46 on: November 05, 2019, 07:03:34 pm »
Do you think that I could pass the 1200 baud 1200Hz/2200Hz through the wired telephone instead, or will I have the same problems as the GSM?
Should work fine. AFAIK, the landline network has the ability to detect modems and change something in the codecs to make them work. Fax machines have relied on this for decades.
Why do people make up rubbish like this?
The possibility of inaccurate recollection (as explained above) is not the same as “making up rubbish”. (And absent documentation either way, I’m not certain that my memory is actually wrong.)

Edit: documentation found, thanks to ollopa remembering the specifics. So yeah, I was right. Definitely not "rubbish", and definitely not "made up". (Also, how much more could I have couched it in uncertainty? I made it clear that I wasn't 100% sure, prefacing it with "AFAIK" and only saying that "something" got changed.)
What ollopa noted is that modems generally signal to disable echo suppressors and echo cancellers. This has nothing to so with the codecs that are used.

 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #47 on: November 05, 2019, 07:13:26 pm »
OK, I see what you’re saying. Sorry I chose the word “codec” and didn’t just leave it at “...changes something in how the PSTN handles calls”. I defer to your infinite knowledge and boundless grace and generosity of spirit in this matter...  ::)
 

Offline sv3oraTopic starter

  • Regular Contributor
  • *
  • Posts: 84
  • Country: gr
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #48 on: November 14, 2019, 01:10:02 am »
Well despite said it can't be done, I have managed to pass 1200 baud AFSK (Bell202) through GSM and also between GSM and landline.
I have made a small page for it with the experiments I did
http://qrp.gr/gsmhack

I want to investigate how valid is this through out the countries.
Would anyone be kind enough to download the file on the page and test it for me in your country's network?

To test this yourself, I have created a zip file in the page, which contains two audio wav recordings. One of them is random FSK data at 1200 baud following the Bell202 tones protocol. The other is a single continuous DTMF tone, the number 2. Set up your audio player so it can open multiple instances and play these files at the same time, at about the same audio volume (enable repeat track so that they play continuously). If you do not have an audio jack connection from your phone (hands free plug) to the PC, you could try acoustically coupling the phone to the PC, although I have not tested this. If you can hear the FSK data uninterrupted on the other phone, then it works for you. Note that I have deliberately introduced a few short pauses in the original FSK audio data, as an indication of which part is played, without looking at the screen.
 
The following users thanked this post: I wanted a rude username

Offline jhpadjustable

  • Frequent Contributor
  • **
  • Posts: 295
  • Country: us
  • Salt 'n' pepper beard
Re: 1200 baud data transfer over audio passband of cellphone. Is that possible?
« Reply #49 on: November 14, 2019, 08:48:12 pm »
Holding the channel open with DTMF! Fine business! :clap:

Have you verified with a proper demodulator yet? What sort of bit error rates are you getting?
"There are more things in heaven and earth, Arduino, than are dreamt of in your philosophy."
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf