P25 Digital Repeater using Motorola Radios

This forum is dedicated to discussions pertaining specifically to the Motorola ASTRO line of radios (those that use VSELP/IMBE/AMBE), including using digital modulation, digital programming, FlashPort upgrades, etc. If you have general questions please use the General or Programming forums.

Moderator: Queue Moderator

Post Reply
USGOVTECH
Posts: 230
Joined: Sat May 04, 2002 3:00 pm

P25 Digital Repeater using Motorola Radios

Post by USGOVTECH »

UPDATE:

It is not necessary to do internal mods to Motorola radios to pass P25 data. I discovered that you can use Pin 5 (Flat Tx audio) instead of adding 10k resistor and injecting audio at TXa test point. You can use Maxtrac, CM SERIES, CDM SERIES,SM SERIES, RADIUS SERIES for all radios. Use pin 11 (discriminator) with a 10-20K variable resistor in line between TX and RX you must ground the unused pin of the resistor (You can use pin 7). I am working on a data validation circuit that will allow for only P25 data to be passed thru the repeater if so desired.

PINOUTS

PROGRAM RX RADIO FOR PIN 8 ACTIVE LOW
ALSO PLACE JU551 TO A

RX RADIO TX RADIO
PIN 8 COR DETECT PIN 3 PUSH TO TALK
PIN 7 GND PIN 7 GND
PIN 11 DISCRIMINATOR /\/\/\10-20K VR PIN 5 FLAT TX AUDIO

ADJUST VR for 2.9 to 3.5 Khz of DATA.
motorola_otaku
Posts: 1854
Joined: Tue Jan 13, 2004 7:03 am

Post by motorola_otaku »

What's the basis for your P25 validation circuit? I've thought about taking one of the Bearcat BC-i25 scanner cards and trying to derive "P25 COR" from it.
User avatar
The Pager Geek
Posts: 1250
Joined: Fri Jun 21, 2002 10:31 pm
What radios do you own?: Disney FRS

Post by The Pager Geek »

I'm going to go out on a limb and say you are doing this with 2 UHF radios? Not crossband....

It should work with a UHF, but a VHF Maxtrac to a UHF maxtrac (Crossband) won't work with this set-up.

tpg
Experienced Provider of Useless Information
USGOVTECH
Posts: 230
Joined: Sat May 04, 2002 3:00 pm

Post by USGOVTECH »

As of yesterday afternoon I have used:

2 VHF GM300's
2 VHF MAXTRACS
1 VHF GM300 and 1 VHF MAXTRAC (TX and RX for Both Radio)

The same configuration for UHF Radios

This morning I just tried 1 VHF GM RX and 1 UHF GM TX and it works fine.
USGOVTECH
Posts: 230
Joined: Sat May 04, 2002 3:00 pm

Post by USGOVTECH »

The validation circuit will be Data detector circuit, it will only validate if digital data is present.
motorola_otaku
Posts: 1854
Joined: Tue Jan 13, 2004 7:03 am

Post by motorola_otaku »

USGOVTECH wrote:The validation circuit will be Data detector circuit, it will only validate if digital data is present.
That's cool, but how? Are you building a circuit from the ground-up, or modifying an existing product?
User avatar
The Pager Geek
Posts: 1250
Joined: Fri Jun 21, 2002 10:31 pm
What radios do you own?: Disney FRS

Post by The Pager Geek »

The VHF Maxtrac to UHF Maxtrac had a problem with one of the receivers inverting the data. (Can't remember which off the top of my head)

I had to either modify the "inverting" receiver, or change where the audio from the Rx's Disc entered the TX Radio.

Just an FYI

tpg
Experienced Provider of Useless Information
USGOVTECH
Posts: 230
Joined: Sat May 04, 2002 3:00 pm

Post by USGOVTECH »

I didnt have to do any mods at all. HMMMMMMMM.
User avatar
wavetar
Administrator
Posts: 7340
Joined: Sun Sep 09, 2001 4:00 pm

Post by wavetar »

Perhaps the GM's were slightly different from the Maxtracs in that respect.

Todd
No trees were harmed in the posting of this message...however an extraordinarily large number of electrons were horribly inconvenienced.

Welcome to the /\/\achine.
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Post by MattSR »

What is the range and reception like compared to a real Quantar?
User avatar
mr.syntrx
Posts: 1587
Joined: Wed Apr 28, 2004 10:09 pm

Post by mr.syntrx »

Quite a bit worse. A real Quantar regenerates the digital signal - this thing just passes it straight through as it hears it, noise included.
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Post by MattSR »

Have you used one of these before or done a back to back comparison?

P25 error detection is pretty damn robust, if a repeater is getting a noisy signal the total BER *should* be the same as if it was working in simplex mode (a la the above repeater). A Quantar can't magically correct uncorrectable errors.

Just interested in how well these things work
User avatar
mr.syntrx
Posts: 1587
Joined: Wed Apr 28, 2004 10:09 pm

Post by mr.syntrx »

Not with P25 specifically, but I've done the same thing with other digital systems.

A Quantar can correct quite a few errors - that's what Forward Error Correction codes are for. When it receives a signal, it will do this, and generate a brand new, clean signal.

This thing, on the other hand, will add noise to what may already be a rather crappy signal if you're trying to hit the machine from the fringes. The signal's journey from the repeater to the receiving radio will then also result in futher deterioration to what may already be a bad signal - whereas from a Quantar, you'd be listening to a fresh, clean, error corrected signal.
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Post by MattSR »

OK, I see what your saying..

I think it would still make for an interesting project though, if it can get 90% of the performance for 10-20% of the cost then id call it a success, especially if you can use DTMF tones to switch a PL on and off and use it as a conventional HAM repeater too.

Of course, nothing beat a Quantar!

Cheers,
Matt
User avatar
Wowbagger
Aeroflex
Posts: 1287
Joined: Tue Dec 10, 2002 10:46 am

Post by Wowbagger »

MattSR wrote:P25 error detection is pretty damn robust, if a repeater is getting a noisy signal the total BER *should* be the same as if it was working in simplex mode (a la the above repeater).
BER, while a useful end-to-end metric of a signal, is NOT the appropriate measurement to be making here - modulation fidelity is.
Yes - a high BER will cause your signal to fall on its face. But what causes a high BER? It is the radio's inability to distinguish between symbols. And what is the measurement of how "hard" symbols are to distinguish? Modulation fidelity. You will see pretty much the same BER for a 10% modulation fidelity error as you will for a 0.7% error, but if you see a 10% MFE you can know the system is NOT performing well - you have lost most of your margin for decode, and any increase at all in the distortion of the signal will start causing errors.

Think of it as the difference between SINAD and intelligibility - you can have a 20 dB SINAD or a 12 dB SINAD, and still understand what the other guy is saying 100% - but if you are at a 12 dB SINAD coming out of the repeater, any additional noise is going to start causing a whole lot of "What did you say? Say again?" traffic.

The advantage of regenerating the digital signal, even if you don't apply the forward error correction at that point, is that you can go from a 10% MFE to a 0.7% MFE, and thus improve the odds of correctly decoding the signal.

AND, if you then apply the FEC to the signal, you can further clean it up, so that you can once again pick up bit errors without the radio being unable to decode.
This is my opinion, not Aeroflex's.

I WILL NOT give you proprietary information. I make too much money to jeopardize my job.

I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
f1shs
Posts: 26
Joined: Wed Jul 12, 2006 8:57 am

Post by f1shs »

I've connected to GM950 together, as described on this post.

it doesn't work :-(

My radio is 12,5 kHz channel. Maybe i've a trouble with this kibg of radio ?

please help.

Pierre
USGOVTECH
Posts: 230
Joined: Sat May 04, 2002 3:00 pm

Post by USGOVTECH »

The deviation out of the Portable or Mobile must remain the same as it is transmitted thru the repeater this is why the audio adjust pot is required.
f1shs
Posts: 26
Joined: Wed Jul 12, 2006 8:57 am

Post by f1shs »

Hi folks,

I've two GM950 model M08RHH4AN4AN connected together with :

RX pin 8 (CSQ detect - low active) =>>> TX pin 3 (PTT - low active)

RX pin 11 (RX Discri) =>>> TX pin 5 (Flat Tx) trough a 22 k variable resistor

RX pin 7 (GND) =>>> TX pin7 (GND)

I'm using two XTS3000 model H09RDH9PW7BN host R07.09.02 DSP 08.02.04 flashcode 500008-000000-1. Those two radios are configured in 2,5 kHz deviation in ASTRO mixed mode C4FM, low power for lab testing.

The repeater is on the table for lab with a dummy load for TX (433.400 Mhz and no antenna for RX (439.800 MHz).

1) in analog mode, the repeater is working fine. The pot is able to adjust the tx audio.

2) in astro mode, i've got nothing

As i ask before, is it a limitation using a 12,5 kHz TX for repeating ASTRO P25 ?

Pierre
USGOVTECH
Posts: 230
Joined: Sat May 04, 2002 3:00 pm

Post by USGOVTECH »

The radios may be limiting the Data. Typically the P25 data is 2.83K, The narrow band radios may not work as well.
f1shs
Posts: 26
Joined: Wed Jul 12, 2006 8:57 am

Post by f1shs »

Ok Thank's USGOVTECH ;-)

I'm gonna use a MC2100 with 25 kHz spacing channels capabalities.

As you told, it's probaly a limitation in the RX bacause i'm using a 12,5 kHz spacing radio (regarding the IF).

Keep'n touch.

Pierre
User avatar
Wowbagger
Aeroflex
Posts: 1287
Joined: Tue Dec 10, 2002 10:46 am

Post by Wowbagger »

Make sure that the audio path has as flat a passband response as possible - take the audio directly from the detector, and route it as directly to the modulator as possible.

Remember that many radios have either a C-weighted audio response, or some other bandpass response. *ANY* non-flat frequency response will KILL the APCO-25 waveform - drive the intersymbol interference through the roof, and make your modulation fidelity error HUGE.

Can you get access to something that can measure P25 modulation fidelity (and it's pretty obvious what I'd recommend...) - if you can actually get something that can show you the eye diagram or modulation fidelity that would really help you identify the problem.
This is my opinion, not Aeroflex's.

I WILL NOT give you proprietary information. I make too much money to jeopardize my job.

I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
f1shs
Posts: 26
Joined: Wed Jul 12, 2006 8:57 am

Post by f1shs »

it work's !!!

I've change the RX radio for a 25 kHz spacing channel capable, and it's works very fine ;-)

We've looked at the TX deviation to be in the limit of 3 kHz.

So i anderstand now that you need a pair of 25 kHz radio to make the P25 repeater become a true story ;-)

many thank's to all of you.

Now we have the first P25 repeater in Paris.

Pierre
USGOVTECH
Posts: 230
Joined: Sat May 04, 2002 3:00 pm

Post by USGOVTECH »

Yes that is correct. Great to hear that it works. The Dev should be around 2.83 khz +/- a few hz.
User avatar
mancow
Batboard $upporter
Posts: 2374
Joined: Sun Sep 09, 2001 4:00 pm

Post by mancow »

I just got one working (in band) Uhf using an HT600 as the RX radio and a Bendix King LPU as the TX unit.


Smooth...... 8)
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

P25 ADP AlgoID

Post by MattSR »

Hi guys,

Does anyone with a P25 service monitor here know the AlgoID for ADP? Ive discovered some common algos in P25 training docs and whatnot, but the value for ADP escapes me, and I dont have a 2975 to check it with :)

Discovered some more AlgIDs (but not ADP sadly)

Heres the values that I have found (from publicly available docs on the net)

0x00 Accordion 1.3
0x01 BATON (Auto even)
0x02 Firefly
0x03 Mayfly
0x04 Saville

0x41 BATON (Auto Odd)

0x80 Clear
0x81 DES-OFB
0x82 2 key Triple DES
0x83 3 key Triple DES
0x84 AES-256

0x9F DES-XL
0xA0 DVI-XL
0xA1 DVP-XL
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Re:

Post by MattSR »

Hi wowbagger,

Its only taken me three years to digest your post - but yes you are quite right, I've learned a fair bit since I made my original post, and thanks for the explanation - it has pointed me i the right direction :)

Cheers,
Matt
Wowbagger wrote:
MattSR wrote:P25 error detection is pretty damn robust, if a repeater is getting a noisy signal the total BER *should* be the same as if it was working in simplex mode (a la the above repeater).
BER, while a useful end-to-end metric of a signal, is NOT the appropriate measurement to be making here - modulation fidelity is.
Yes - a high BER will cause your signal to fall on its face. But what causes a high BER? It is the radio's inability to distinguish between symbols. And what is the measurement of how "hard" symbols are to distinguish? Modulation fidelity. You will see pretty much the same BER for a 10% modulation fidelity error as you will for a 0.7% error, but if you see a 10% MFE you can know the system is NOT performing well - you have lost most of your margin for decode, and any increase at all in the distortion of the signal will start causing errors.

Think of it as the difference between SINAD and intelligibility - you can have a 20 dB SINAD or a 12 dB SINAD, and still understand what the other guy is saying 100% - but if you are at a 12 dB SINAD coming out of the repeater, any additional noise is going to start causing a whole lot of "What did you say? Say again?" traffic.

The advantage of regenerating the digital signal, even if you don't apply the forward error correction at that point, is that you can go from a 10% MFE to a 0.7% MFE, and thus improve the odds of correctly decoding the signal.

AND, if you then apply the FEC to the signal, you can further clean it up, so that you can once again pick up bit errors without the radio being unable to decode.
newtomotorola
Posts: 116
Joined: Wed Jan 23, 2008 6:27 am

Re: P25 Digital Repeater using Motorola Radios

Post by newtomotorola »

Sorry to digg up an old thread, but what are people's recommendations on UHF versions that would do this say around the 420-450 area

Thanks
User avatar
Astro Spectra
Posts: 668
Joined: Sat Sep 22, 2001 4:00 pm

Re: P25 Digital Repeater using Motorola Radios

Post by Astro Spectra »

The $64,000 question was "What's the basis for your P25 validation circuit?". Well, perhaps the answer might be found here:

http://nicta.com.au/__data/assets/pdf_f ... ware_1.pdf

Now before anyone starts going on about DVSI (and other's) patents notice that this project was to convert P25 RF received off air to digital data displayed on PC and it wasn't intended as a saleable product.

Well it turns out that this project was based on the GNUradio free software development signal processing toolkit. You can do further reading here:

http://sedition.org.au/op25/wiki/WikiStart

and

http://gnuradio.org/trac/wiki

plus a cool 4FSK software demod plug in from:

http://radiorausch.googlepages.com/Gnur ... elFSK.html

BUT and it's a big BUT, this is a major major project with good C++ skills required. It is hardly ‘batteries not included’, it’s not ‘some assembly required’, and it’s certainly not 'plug and play' like Unitrunker.

Well actually a whole lot of assembly is required and while it looks like the transmit functions are still a bit primitive, these Australians look pretty serious.

Note also that the implementation is done via sampling either the RF or IF rather than by means of a discriminator tap (although that is certainly possible). The whole software-defined radio thing looks a bit overly complex and expensive when using the typical USRP front end when most people here would be able to lay their hands on suitable RF decks but Max KA1RBI has done a simple 455 kHz to sound card IF design is here:

http://www.lightlink.com/mhp/iq/

So the whole "P25 validation circuit" is going to look like a small PC server.

Oh, and here's the thing - you're not going to be trying to sell this as a commercial product because while you don't need an IMBE decoder for a repeater you will still be using all sorts of IP that belongs to others...
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Re: P25 Digital Repeater using Motorola Radios

Post by MattSR »

We are very serious - this is part of Steve Glass's Comp Sci Phd thesis, so we clearly aren't mucking around.

At the moment testing is underway on pure I/Q decoding as opposed to Franks C4FM slicer.
Astro Spectra wrote:So the whole "P25 validation circuit" is going to look like a small PC server.

Oh, and here's the thing - you're not going to be trying to sell this as a commercial product because while you don't need an IMBE decoder for a repeater you will still be using all sorts of IP that belongs to others...
Why? without IMBE I cant see anything in a P25 repeater that's proprietary or IP...
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Re: P25 ADP AlgoID

Post by MattSR »

Well. I finally found it - I've succesfully captured the data using the OP25 project and GNUradio framework and it all decodes perfectly!

The AlgID is 0xAA.

Its sad that 90% of the people that look at this post won't understand the significance of what has been achieved with this project, but here it is anyway :)

Image
Image
Image
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Re: P25 Digital Repeater using Motorola Radios

Post by MattSR »

Thanks to Max, KA1RBI, some more progress has been made on OP25.

I used a Ubuntu running inside VMware, and a disc tapped uniden scanner to create the following screenshots:-

The first one shows a grab of an ADP transmission, the second shows the output of the C4FM slicer (work continues on a I/Q decoder that handles both CQPSK and C4FM)

Image
Image
Image

Cheers,
Matt
User avatar
Wowbagger
Aeroflex
Posts: 1287
Joined: Tue Dec 10, 2002 10:46 am

Iiiinnnntersssting!

Post by Wowbagger »

Very interesting. Thanks for posting this.
This is my opinion, not Aeroflex's.

I WILL NOT give you proprietary information. I make too much money to jeopardize my job.

I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Re: P25 ADP AlgoID

Post by MattSR »

No worries - All that was used is Ubuntu running in VMware, and a discriminator tapped Uniden scanner..

There are other things coming too :)
User avatar
escomm
Queue Moderator
Posts: 5170
Joined: Fri Mar 24, 2006 8:24 pm

Re: P25 ADP AlgoID

Post by escomm »

Cool
User avatar
The Pager Geek
Posts: 1250
Joined: Fri Jun 21, 2002 10:31 pm
What radios do you own?: Disney FRS

Re: P25 Digital Repeater using Motorola Radios

Post by The Pager Geek »

Lookin Sweet
Experienced Provider of Useless Information
newtomotorola
Posts: 116
Joined: Wed Jan 23, 2008 6:27 am

Re: P25 ADP AlgoID

Post by newtomotorola »

Hi Matt, mabie you could post up a step by step on how to get it working through vmware. I think what you guys are doing is fantastic, keep up the good work.

Cheers
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Re: P25 ADP AlgoID

Post by MattSR »

No worries, I need to update the wiki with some doco on how to do it.

I used VMware Fusion on a Mac, though it shouldn't be too hard to get it running on other platforms.
newtomotorola
Posts: 116
Joined: Wed Jan 23, 2008 6:27 am

Re: P25 Digital Repeater using Motorola Radios

Post by newtomotorola »

Thanks mate highly appreciate it, would love to get it going and having abit of a play.
MattSR
Posts: 770
Joined: Mon Apr 21, 2003 10:00 pm

Re: P25 Digital Repeater using Motorola Radios

Post by MattSR »

Back on the topic, transmit code has been added to OP25. One step closer to a software defined repeater. :)
User avatar
kb9suy
Posts: 311
Joined: Fri Dec 21, 2001 4:00 pm
What radios do you own?: too many!!

Re: P25 Digital Repeater using Motorola Radios

Post by kb9suy »

Has anyone been able to do a cross band with gm300's succesfully? I can pass VHF to UHF Can not pass UHF to VHF any luck.
Post Reply

Return to “Legacy Batboard Motorola ASTRO (VSELP/IMBE/AMBE) Equipment Forum”