Who's REALLY good at hacking DSP? Idea...
Moderator: Queue Moderator
- Elroy Jetson
- Posts: 1158
- Joined: Mon Sep 03, 2001 4:00 pm
Who's REALLY good at hacking DSP? Idea...
The Astro line (at least, the Astro Saber) uses the Motorola 56001 DSP chip. Because of the use of that chip, and (apparently) the rest of the radio's
topology as well, it's THEORETICALLY capable of operating in AM mode.
There are various bits and pieces of code for the 56001 chip out there that you can download which are used for some AM applications including
AM stereo decoding. There's no doubt that the hardware is AM capable. (Being IQ based, most modulation schemes are possible.)
I'm no coder. But is there anyone here who might be able to investigate the AM possibilities for Astro radios?
There's just one application that's really meaningful for AM on an Astro, and that's to listen to aircraft band transmissions below the VHF band, which
by itself takes another software hack.
I guess it would be a pretty complex project, but it would be pretty cool to have an Astro operating in AM mode.
Any takers?
Elroy
topology as well, it's THEORETICALLY capable of operating in AM mode.
There are various bits and pieces of code for the 56001 chip out there that you can download which are used for some AM applications including
AM stereo decoding. There's no doubt that the hardware is AM capable. (Being IQ based, most modulation schemes are possible.)
I'm no coder. But is there anyone here who might be able to investigate the AM possibilities for Astro radios?
There's just one application that's really meaningful for AM on an Astro, and that's to listen to aircraft band transmissions below the VHF band, which
by itself takes another software hack.
I guess it would be a pretty complex project, but it would be pretty cool to have an Astro operating in AM mode.
Any takers?
Elroy
It isn't that simple, I'm afraid
It isn't quite as simple at that.
First, on the RX path: you have the AGC loops in the analog section prior to the digitizer for the IF. They are going to try to flatten out any RF level variations in the received signal - that is, after all, their job in life. As a result they will try to flatten out any AM they see. To be able to receive AM, you have to have an AGC loop response time slower than about 3 msec (about 300 Hz). However, that slow an AGC response time will bite you when you try to deal with a bursty signal. Mobiles don't need to deal with bursty signals as much as the base radio does, but still - the if AGC loops run too fast, you are not going to get the information to the ADC.
You have a similar problem on the TX path (though for your use case this doesn't matter). If the AGC that controls the RF level runs too fast, it will squish any AM on the signal.
First, on the RX path: you have the AGC loops in the analog section prior to the digitizer for the IF. They are going to try to flatten out any RF level variations in the received signal - that is, after all, their job in life. As a result they will try to flatten out any AM they see. To be able to receive AM, you have to have an AGC loop response time slower than about 3 msec (about 300 Hz). However, that slow an AGC response time will bite you when you try to deal with a bursty signal. Mobiles don't need to deal with bursty signals as much as the base radio does, but still - the if AGC loops run too fast, you are not going to get the information to the ADC.
You have a similar problem on the TX path (though for your use case this doesn't matter). If the AGC that controls the RF level runs too fast, it will squish any AM on the signal.
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.
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.
- Elroy Jetson
- Posts: 1158
- Joined: Mon Sep 03, 2001 4:00 pm
Re: Who's REALLY good at hacking DSP? Idea...
Good points, but I've hacked the Astro software to extend the VHF range down as far as the VCO will lock, before, and it was able to receive AM
aircraft transmissions with a usable degree of intelligibility most of the time. The precise tuning of the signal is important. The FM detector has to
detect "on the slope" in order to pull out the AM information. Signals that were just right were 100 percent intelligible. So I don't think the AGC is
really going to be a fatal problem.
Elroy
aircraft transmissions with a usable degree of intelligibility most of the time. The precise tuning of the signal is important. The FM detector has to
detect "on the slope" in order to pull out the AM information. Signals that were just right were 100 percent intelligible. So I don't think the AGC is
really going to be a fatal problem.
Elroy
- Astro Spectra
- Posts: 669
- Joined: Sat Sep 22, 2001 4:00 pm
Re: Who's REALLY good at hacking DSP? Idea...
Elroy,
I've not taken the time to look at the schematics, but if Wowbagger is right and there is an AGC loop to keep the input to the ABACUS IC within the range of the A/D then it might be way easier to take a sample of the AGC (ie a piece of wire) and couple it to one of the audio stages. The AGC loop might have a quite a low pass filter in it so you'll want to pick off close to the detector.
However in the back of my mind is the fact the A/D might be a delta sigma converter doing its own AGC detecting. I really need a look at that schematic....
Now the mute is not a problem as the CSQ noise mute action will work fine for AM (many commercial AM rigs implement the mute this way). What you'll need to find is a cunning way to switch the audio stage from the audio D/A output to the AC coupled AGC line, for example if you're not using secure you could use the secure rotary.
John
I've not taken the time to look at the schematics, but if Wowbagger is right and there is an AGC loop to keep the input to the ABACUS IC within the range of the A/D then it might be way easier to take a sample of the AGC (ie a piece of wire) and couple it to one of the audio stages. The AGC loop might have a quite a low pass filter in it so you'll want to pick off close to the detector.
However in the back of my mind is the fact the A/D might be a delta sigma converter doing its own AGC detecting. I really need a look at that schematic....
Now the mute is not a problem as the CSQ noise mute action will work fine for AM (many commercial AM rigs implement the mute this way). What you'll need to find is a cunning way to switch the audio stage from the audio D/A output to the AC coupled AGC line, for example if you're not using secure you could use the secure rotary.
John
Re: Who's REALLY good at hacking DSP? Idea...
I've seen an AstroSaber3 VHF that was AM. Even had a Red display. I'm not sure at all about how it was done, but it exists and I guess it's possible. I didn't look on the inside at all so who knows what was in there..... but it's out there.
- Elroy Jetson
- Posts: 1158
- Joined: Mon Sep 03, 2001 4:00 pm
Re: Who's REALLY good at hacking DSP? Idea...
A hack to extend the VHF band limit down to the aircraft band is very easy. I've done it by the usual bandsplit hack method.
And it worked, to a point, but it was not true AM, it was just doing "slope detection" which is not a perfect way of demodding AM but will
yield usable recovered audio IF the radio is tuned exactly right. But it's not true AM reception. And the radio will still be an FM radio
in every way. And the red display is easy enough. A few suitable surface mount LEDs in the desired color are easy enough to install
in the display module.
If it was a TRUE AM mod, I'd like to see it.
Elroy
And it worked, to a point, but it was not true AM, it was just doing "slope detection" which is not a perfect way of demodding AM but will
yield usable recovered audio IF the radio is tuned exactly right. But it's not true AM reception. And the radio will still be an FM radio
in every way. And the red display is easy enough. A few suitable surface mount LEDs in the desired color are easy enough to install
in the display module.
If it was a TRUE AM mod, I'd like to see it.
Elroy
Re: Who's REALLY good at hacking DSP? Idea...
Jeebus, Ray... you're gonna be the next Cowthief if you keep going. Did it do Fascinator too?Rayjk110 wrote:I've seen an AstroSaber3 VHF that was AM. Even had a Red display. I'm not sure at all about how it was done, but it exists and I guess it's possible. I didn't look on the inside at all so who knows what was in there..... but it's out there.
Put up or shut up.
-
- Posts: 930
- Joined: Fri Jun 23, 2006 11:21 am
Re: Who's REALLY good at hacking DSP? Idea...
I heard it was FPP too!!!tvsjr wrote:Jeebus, Ray... you're gonna be the next Cowthief if you keep going. Did it do Fascinator too?Rayjk110 wrote:I've seen an AstroSaber3 VHF that was AM. Even had a Red display. I'm not sure at all about how it was done, but it exists and I guess it's possible. I didn't look on the inside at all so who knows what was in there..... but it's out there.
Put up or shut up.
Sorry, couldn't resist.
Re: Who's REALLY good at hacking DSP? Idea...
Yup... I wonder if there's a flashcode option for NAV/VORTAC fixes?Batwings21 wrote:I heard it was FPP too!!!tvsjr wrote:Jeebus, Ray... you're gonna be the next Cowthief if you keep going. Did it do Fascinator too?Rayjk110 wrote:I've seen an AstroSaber3 VHF that was AM. Even had a Red display. I'm not sure at all about how it was done, but it exists and I guess it's possible. I didn't look on the inside at all so who knows what was in there..... but it's out there.
Put up or shut up.
Sorry, couldn't resist.
Re: Who's REALLY good at hacking DSP? Idea...
Yeah; and I saw one that was tri-band too! VHF, UHF, AND 800! The really col thing was, that not only was it tri-band; and have FPP; you could actually change the three bands to whatever you wanted! You could do low band and CB on one of the personalities, then change it through the FPP to 1.2 GHz, and so forth. AND; get this! The antenna was only 3" and had 9db gain on all bands. The radio is smaller than a pack of playing cards! I saw one! I swear! I really saw one! Somebody PLEASE beleive me! Oh; and did I mention that it was also a cell phone, and MP3 player with a color screen and GPS built in it? Also full broadband internet capanilities.
Ok; OK; I'm choking on my own stench after that one...
Ok; OK; I'm choking on my own stench after that one...
Re: Who's REALLY good at hacking DSP? Idea...
That was the one with the blue back light and had the back lit saber keys IIRC. I think it may have been a translucent case too.4n6inv wrote:Yeah; and I saw one that was tri-band too! VHF, UHF, AND 800!
The Radio Information Board: http://www.radioinfoboard.com
Your source for information on: Harris/Ma-Comm/EFJ/RELM/Kenwood/ICOM/Thales, equipment.
Your source for information on: Harris/Ma-Comm/EFJ/RELM/Kenwood/ICOM/Thales, equipment.
Re: Who's REALLY good at hacking DSP? Idea...
Alex,
That's the one!!! You could even keep small fish (probably Neon Tetras) inside the translucent housing. How cool is that? Its also rumored that Motorola is going to come out with one in 2007 Q-4 that has tiny solar AND lunar cells on the top of the case to keep the battery charged at all times. From what I hear from my sources inside Motorola; they are also developing one that has a built in breathalyzer, as well.
I'll keep everyone posted on the progress. See! I told you guys it was true!
OK; enough, already...
That's the one!!! You could even keep small fish (probably Neon Tetras) inside the translucent housing. How cool is that? Its also rumored that Motorola is going to come out with one in 2007 Q-4 that has tiny solar AND lunar cells on the top of the case to keep the battery charged at all times. From what I hear from my sources inside Motorola; they are also developing one that has a built in breathalyzer, as well.
I'll keep everyone posted on the progress. See! I told you guys it was true!
OK; enough, already...
- Astro Spectra
- Posts: 669
- Joined: Sat Sep 22, 2001 4:00 pm
Re: Who's REALLY good at hacking DSP? Idea...
Getting back on to the original topic...
Bad news on the hardware hack Elroy. Looking at the schematic there is no hardware AGC around the IF before the ABACUS and no obvious analog RSSI or other amplitude related analog signal available.
Bad news on the hardware hack Elroy. Looking at the schematic there is no hardware AGC around the IF before the ABACUS and no obvious analog RSSI or other amplitude related analog signal available.
Re: Who's REALLY good at hacking DSP? Idea...
Actually, that would be GOOD news: if there is no AGC from the RF to the ADC, then the ADC has all the information to do AM - assuming the system isn't running fully compressed to begin with. If so, then doing AM demod is simply a matter of taking the digitized IF, doing your final bandwidth limiting, then computing the magnitude of the signal: you can even skip the I/Q generation if that's being done in software (I don't know what the ABACUS chip looks like, so I don't know if it is doing all the functions of a programmable down converter or if it is just digitizing the IF and letting the DSP do all the work - we use PDCs which generate the I/Q baseband signal from the digitized IF).Astro Spectra wrote:Getting back on to the original topic...
Bad news on the hardware hack Elroy. Looking at the schematic there is no hardware AGC around the IF before the ABACUS and no obvious analog RSSI or other amplitude related analog signal available.
If you have I & Q, you just compute I*I-Q*Q and take the square root, and you have magnitude: subtract 0.5 from that (assuming it is scaled to be 0..1) and you have the AM signal.
You can even do SSB if you have I & Q and aren't getting limited in the RF section: run Q through a Hilbert filter, and then add or subtract from I to get USB or LSB.
(Yes, I've actually done this kind of coding in the 2975.)
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.
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.
Re: Who's REALLY good at hacking DSP? Idea...
Note: I removed a large number of funny jab's at Ray from here, please keep on topic.
I don't mind subtle jokes here and there, as long as they are kept friendly... (Ray, your last one crossed the line by far.).
Keep this on topic and in this millennium, and we'll be all good.
-Alex
I don't mind subtle jokes here and there, as long as they are kept friendly... (Ray, your last one crossed the line by far.).
Keep this on topic and in this millennium, and we'll be all good.
-Alex
The Radio Information Board: http://www.radioinfoboard.com
Your source for information on: Harris/Ma-Comm/EFJ/RELM/Kenwood/ICOM/Thales, equipment.
Your source for information on: Harris/Ma-Comm/EFJ/RELM/Kenwood/ICOM/Thales, equipment.
Re: Who's REALLY good at hacking DSP? Idea...
All I was simply trying to state is that I've seen it, and I've seen it work with my own 2 eyes, so I'm just clarifying to the OP that the idea is indeed feasable. How? I do not know. But that's half the fun with these things is figuring out!
Sorry if I upset you, Terry.

Sorry if I upset you, Terry.