Wideband digital transmissions - monitoring?

[kb1ipd]

There is an issue which I see becoming a big problem for general purpose monitoring of communications: The increased use of wide-band digital transmissions for mobile, broadcast and other uses in communications. To decode a digital signal, one needs to sample it and get the information into a computer or decoding device. However, many of these signals far exceed the bandwidth of a sound card and are therefore going to be difficult to decode by any conventional method.

(Note, I've included some which are likely to be encrypted and would not be legal to decode, but that's just for reference


GSM: ~200 kHz
IS-95 (CDMA): 1.23 MHz
CDMA2000 1xRTT: 1.25 MHz
CDMA2000 3xRTT 3.75 MHz
CDMA EVDO: 1.25-5.0 Mhz
UMTS (WCDMA): 5.0 Mhz
WiBro: 8.75 MHz (maximum per-channel)
Wimax: 3.5 MHz, 5 MHz, 7 MHz, 8.75 Mhz, 10 Mhz (per channel, depending on profile)
Wifi: 22Mhz Per Channel
Bluetooth: ~1Mhz



Digital Video (Broadcast and otherwise):

Dvb-T: 6, 7, 8 Mhz
ISBD: 6,7, 8Mhz
DVB-T2: 1.7, 5, 6, 7, 8, and 10 MHz
DVB-H: Up to 8 Mhz
ATSC: 6 Mhz



Other:

DAB: 1.712Mhz Max



Of course it should be mentioned that many of these signals do not occupy the entire channel at once, but instead use multiple access systems or spread spectrum of some type. That being the case, they're not necessarily that difficult to decode with the correct equipment to track and receive the signal. However, the issue with such an approach is that it would necessitate a different dedicated hardware receiver for each and every data mode.


I've looked at ways of getting such signals into a computer and it seems options are limited. There are some ADC's that can do it, but they don't seem to be geared toward radio monitoring and thus they may not have any software that can handle these kinds of signals.

Of course, there's always software defined radios, but as of yet, most I've seen don't have adiquate bandwifth to do that kind of sampling. Even those that do present the other problem of signal processing. While software defined radios seem great in so far as their flexibility, it seems that many tasks that are comperably easy in hardware are extremely processor intensive in software. For example, demodulating and decoding a simple ATSC HDTV broadcast entirely in software, from the signal is so processor intensive that even high end computers have trouble doing it in real time.

Granted, computers will get faster given time, but signals may become even more wideband as time goes on. Also, even given the current increase in computer power, we may be a good few years away from the point where an affordable home computer can do something like demodulate ATSC in real time without breaking a sweat.

Is this going to mean an end to monitoring as we know or knew it? I hope not

[N_Jay]

There has never been "general purpose monitoring of communications", there has just been an overwhelming number of services sharing a common transmission mode.

Cellular, and generally not monitor able (and not legal in the US)
GSM: ~200 kHz
IS-95 (CDMA): 1.23 MHz
CDMA2000 1xRTT: 1.25 MHz
CDMA2000 3xRTT 3.75 MHz
CDMA EVDO: 1.25-5.0 Mhz
UMTS (WCDMA): 5.0 Mhz

Datacomm, and generally monitoring/hacking datacomm networks is not desired, and not legal many places.
WiBro: 8.75 MHz (maximum per-channel)
Wimax: 3.5 MHz, 5 MHz, 7 MHz, 8.75 Mhz, 10 Mhz (per channel, depending on profile)
Wifi: 22Mhz Per Channel
Bluetooth: ~1Mhz

Broadcast or subscription; Either get the correct receiver, and sign up is subscription.
Dvb-T: 6, 7, 8 Mhz
ISBD: 6,7, 8Mhz
DVB-T2: 1.7, 5, 6, 7, 8, and 10 MHz
DVB-H: Up to 8 Mhz
ATSC: 6 Mhz
DAB: 1.712Mhz Max

P.S. None of this has anything to do with Digital Trunking

[kb1ipd]

Originally Posted by N_Jay

P.S. None of this has anything to do with Digital Trunking

Well there was no other place that seemed totally suited for digital so I figured this was as good as any.


I'm talking about the fact that there was a time (not long ago) when if something was a data signal, with the exception of perhaos wideband satellite and such things, it was the kind of thing that you could sample on a computer and then either decode or at least observe. If you were picking up a digital mode of unknown origin there are multimode software packages that can decode Pactor, APRS, BAUDOT, wefax, Pactor II. ACARS etc.

It is becoming increasingly impossible to even sample the newer modes and I think that the future is going to be more and more oriented toward digital data of modes that are far too wideband to monitor at all or will require a dedicated receiver for each one. If the mode is used on a non-standard frequency or something then you're out of luck.

[N_Jay]

Originally Posted by kb1ipd

. . . It is becoming increasingly impossible to even sample the newer modes and I think that the future is going to be more and more oriented toward digital data of modes that are far too wideband to monitor at all or will require a dedicated receiver for each one. If the mode is used on a non-standard frequency or something then you're out of luck.

And I have a whole set of those really short screwdrivers for adjusting carburetors that are completely useless on fuel injection.

Technology moves on! Get used to it.

[EnidPuceflange]

Actually, viewed from a system perspective, these examples you give are a great indication of convergance - they all use wideband transmission, and some form of digital IF receiver. Since computers have increasing power, and since these modulation formats are all described forms of I&Q modulation it is clear that there will be a software-based receiver for many if not all of these in the near future. You might want to google the term "Software defined radio" for some hints as to where this is going.

As N_Jay says, technology marches on....

[N_Jay]

Originally Posted by EnidPuceflange

Actually, viewed from a system perspective, these examples you give are a great indication of convergance - they all use wideband transmission, and some form of digital IF receiver. Since computers have increasing power, and since these modulation formats are all described forms of I&Q modulation it is clear that there will be a software-based receiver for many if not all of these in the near future. You might want to google the term "Software defined radio" for some hints as to where this is going.

As N_Jay says, technology marches on....

Careful putting the word "soon" in the same post as "SDR", when talking about convergence.

[KA1RBI]

Originally Posted by kb1ipd

I've looked at ways of getting such signals into a computer and it seems options are limited. There are some ADC's that can do it, but they don't seem to be geared toward radio monitoring and thus they may not have any software that can handle these kinds of signals.

One thing which I've yet to see mentioned in this thread is the USRP / USRP2 (ettus.com) . It's available today (although I haven't saved up enough of the pennies yet to buy one)...

There are plug-in receiver modules available for various bands. Best part is that all of the needed software is Free Software. We already have P25 decoding working [ see sedition.org.au ]. Another very cool site with a lot of applications including some wideband stuff is the Radio Rausch site on googlepages.com. [ radiorausch - Frank's Fantastically Fabulousradiorausch ]

Also, I've done some experiments using a TV tuner as a front end. The UHF one that I tried had a very bad oscillator, which completely ruined any chance of narrow-band work. I.e., you feed a sine wave in at 900 MHz, get a broad band noise burr out at the CH3-4 IF.

If this could be fixed up, a downconverter would be the next step. Many soundcards now run faster than 48,000 / sec. - a 96K soundcard would give you a 48 KHz slice worth of bandwidth to play with....

73

Max

P.S. OK, after previewing it appears that this board has changed the URL's which I originally pasted into the above. Hopefully they haven't been mangled..........

[N_Jay]

Originally Posted by KA1RBI

. . . . We already have P25 decoding working [ see sedition.org.au ]. . . ........

It does not look like they have the vocoder implemented (unless I misread something or missed something)

[KA1RBI]

Originally Posted by N_Jay

It does not look like they have the vocoder implemented (unless I misread something or missed something)

Well, new members would be most welcome to join the project and help out by contributing. You need not be a programmer - also very useful would be testing and documentation. The project has a mailing list on yahoo-groups named op25-dev.

That said, Stevie (the project leader) *does* have at least the IMBE decoder code in working condition now. I don't know about the voice encoder. He has quietly offered to send the source code to anyone who asks [ politely, I'd suggest ] At this stage the overall state of the project sort of limits its use to geeks who would know what to do upon receiving the source code. This is where I reiterate the point that new project members would be welcome to help change that situation by helping to make everything more accessible to more people. I invite anyone who's interested to do so - post your experiences here on R.R. as well...

73

Max

[N_Jay]

Originally Posted by KA1RBI

Well, new members would be most welcome to join the project and help out by contributing. You need not be a programmer - also very useful would be testing and documentation. The project has a mailing list on yahoo-groups named op25-dev.

That said, Stevie (the project leader) *does* have at least the IMBE decoder code in working condition now. I don't know about the voice encoder. He has quietly offered to send the source code to anyone who asks [ politely, I'd suggest ] At this stage the overall state of the project sort of limits its use to geeks who would know what to do upon receiving the source code. This is where I reiterate the point that new project members would be welcome to help change that situation by helping to make everything more accessible to more people. I invite anyone who's interested to do so - post your experiences here on R.R. as well...

73

Max

I just enjoy watching the show from the side lines.

As an example, the "OpenP25" group says: "Our goal is to develop an open-source reference implementation of a P25 trunked radio system controller and ISSI switch."
showing that they don't even grasp that there is no ISSI switch.