Provoice listening

[John1984]

I'm located in Tasmania, Australia. The only scanner that can really work or most people down here use is the UBCD396XT. The analogue sounds great. I'm using the really long antennas approx 3 feet that came with the rtl-sdrs.

I'm not sure if the errors occur while it is decoding, I will have to check. Is it normal for the oscilloscope to start making a nice voice signal about 2 to 3 seconds before you here anything?

I think the fun part is getting it all going and figuring out how it works aggie. I've got a long way to go yet!! Is it normal for the decoded voice to be not as clear as the analogue?

I was thinking, and I don't know if this would work.. of buying a Harris radio and setting that up as a scanner. Is that even possible?


Cheers!!

Sent from my SM-G920I using Tapatalk

 

[dave3825]

I'm located in Tasmania, Australia. The only scanner that can really work or most people down here use is the UBCD396XT. The analogue sounds great. I'm using the really long antennas approx 3 feet that came with the rtl-sdrs.

I'm not sure if the errors occur while it is decoding, I will have to check. Is it normal for the oscilloscope to start making a nice voice signal about 2 to 3 seconds before you here anything?

I think the fun part is getting it all going and figuring out how it works aggie. I've got a long way to go yet!! Is it normal for the decoded voice to be not as clear as the analogue?

I was thinking, and I don't know if this would work.. of buying a Harris radio and setting that up as a scanner. Is that even possible?


Cheers!!

Sent from my SM-G920I using Tapatalk

The fact that you are using the antenna that came with the dongle vs an antenna tuned for the freq range you are trying to receive might contribute to the errors.

 

[John1984]

Thanks Dave, I'd like to purchase an antenna tuned for my frequencies. How much do you think I would need to spend?

Sent from my SM-G920I using Tapatalk

 

[AggieCon]

Hello. Im new to all of this. New to scanning, new to sdr, new to this thread. Complete noob. I got two SDR's. I've played in SDR#, and now I've got a couple of trunking systems monitored. Analog is easy peazy, but the provoice has been giving me absolute hell. Sometimes the voice is good, sometimes its really garbled and distorted, and everything in between, and DSD+ shows lots of errors throughout sometimes as screen captured earlier in the thread. Still using the crap antennas that came with the SDR's. I've been thinking (hoping) its just the antenna setup but I think I might have something wrong in unitrunker as well. Please feel free to correct me, but could the bandwidth being 12.5mhz on each frequency be the problem? It was the default and I just left it there, but in SDR# on analog channels if the bandwidth is too narrow it cuts out part of the transmission. Again, take pity on me in your responses, I am a stupid noob

Rctelles, welcome, your post almost slipped through the cracks.. I think you might still be under the moderator approval delay. Getting as far as you describe is a great job!

When you are monitoring trunked systems, can you explain your configuration? Where does SDR# fit in? I usually use just UniTrunker piped to DSD+ with a virtual audio cable. If you are getting any voice at all, I bet it's configured properly. In UniTrunker, on your Voice VCO, digital output should be the virtual audio cable that routes to DSD+ and the "Provice" box under voice should be selected. So you can still catch analog, keep the analog box check and the "Audio Output" routed to your speakers.

My post previous to yours covers some strategies for tacking poor DSD+ decodes. How far away from the transmit site are you, and what antenna are you using? What system are you monitoring? Is it simulcast?

You can also test through SDR# by piping the audio to DSD+ and staying on one of the voice channels (be sure to turn off all of the audio filters); however, unless your system is busy with a lot of ProVoice traffic, it could take some time for something to roll around. Regarding the bandwidth, I usually adjust mine down to a little outside of the signal on the waterfall view. 12.5kHz, which is the narrow banding bandwidth is cutting it pretty close. If your not dead center, you may be missing stuff. I'd widen it. ~20000Hz is usually what I use.

Start by testing gain setting in SDR+ (where you can visualize what's going on). Then adjust the UniTrunker gain appropriately (UniTrunker gain is ~ 10x the SDR+ gain). IN SDR+, what is the best signal peak to noise floor you can achieve? Also, use SDR# to figure out best antenna placement and the best antenna you have available to you.

Once you go through the basics of achieving the best signal reception possible, there are a few tweaks in DSD+ to fine tune the decode. Your biggest improvement, however, will be in the quality of signal you provide DSD+.

 

[AggieCon]

I'm using the really long antennas approx 3 feet that came with the rtl-sdrs.

I'm not sure if the errors occur while it is decoding, I will have to check. Is it normal for the oscilloscope to start making a nice voice signal about 2 to 3 seconds before you here anything?

I think the fun part is getting it all going and figuring out how it works aggie. I've got a long way to go yet!! Is it normal for the decoded voice to be not as clear as the analogue?

I was thinking, and I don't know if this would work.. of buying a Harris radio and setting that up as a scanner. Is that even possible?

I'd like to purchase an antenna tuned for my frequencies. How much do you think I would need to spend?

Howdy John,

It's outside my experience to know if you can get your own Harris radio configured. You'd have to figure out how you would get it programmed. I am not certain the permissions given to the radios on an EDACS system. I know "join" commands are issued over the control channel. I'm not sure if one of their handhelds has to comply (obviously yours would have to be modified to NEVER transmit). At any rate, before you get too excited about that idea, EDACS is end of life at the end of this year. Harris will no longer support it. So you'd want to see how long the system you monitor will stick around, or if they have plans already to replace it. If you do figure all of that out, though, I'd be very interested to hear about the project. You might search the forum to see if others have had success.

When someone wants to talk, they hit the PTT. It is transmitted over the control channel uplink. If the talk group is free and there is an open voice channel, the computer tells all of the radios on the talkgroup to go to the voice channel. Once on the voice channel, the calling unit sends additional data to verify to the computer and other radios that everything is set to transmit. The other radios then open the squelch, and the user begins his conversation. All of this takes a bit of time.

On our side, when UniTrunker gets the command from the control channel, it retunes the voice receiver to take the call. So from an audio output perspective, at the beginning you get a bit of whatever the tuner was parked at, you get silence while it retunes, you get the sync data (from the transmitting unit) and then you get the voice traffic. Also, they might wait a second or two to start talking. At the end, you will get static noise (UniTrunker holds on to calls too long IMO) and then silence (unless you have the voice receiver parked and no squelch setting).

Below are a few frame grabs of the process. Before you get to my point, you will likely see static, silence, and then the first picture I have, which I think is the sync data. The middle picture is the meat: This is the encoded voice data. Finally, the last picture is the static sent once they are done with the voice channel yet UniTrunker holds on to it.

I record my traffic, and my recordings usually have ~1 second of silence before they start talking. I don't know what may cause 2-3 seconds of silence. Do you think you are missing the beginning of the conversation?

To me, the ProVoice traffic is not as "rich" as the analog traffic. This is due to the encoding process. Only sound frequencies below ~4kHz are encoded. Additional filters are used to modify the input in a way that they though made it easier to hear. So the Provoice traffic will be duller than the analog. P25 voice encoding is MUCH better. However, if by "clear," you mean glitchy, like the audio cuts out, starts and stops, that's likely your decode quality (I say likely because end users have this problem from time to time as well).

Here is a sample of ProVoice I just recorded.

Another problem is sometimes they have their input (analog to digital converter) gain too high, which causes the voice to be distorted, like the very first call in the recorder.

Regarding an antenna, what system are you trying to monitor? What are the frequencies? How far away are you? Is this going inside or outside? At what height? There are a lot of considerations for getting an antenna that matches what you are trying to do.

By the way, thanks for the compliment, earlier, that made my day.
​

 

[slicerwizard]

Looks like there is something off with your setup. The higher frequencies are being attenuated.

 

[rctelles]

When you are monitoring trunked systems, can you explain your configuration? Where does SDR# fit in? I usually use just UniTrunker piped to DSD+ with a virtual audio cable. If you are getting any voice at all, I bet it's configured properly. In UniTrunker, on your Voice VCO, digital output should be the virtual audio cable that routes to DSD+ and the "Provice" box under voice should be selected. So you can still catch analog, keep the analog box check and the "Audio Output" routed to your speakers.

SDR# doesnt fit in. What you describe is how I have it set up, except the analog passthrough. I wasnt aware that was possible. I meant that in SDR# the 12.5mhz bandwidth looks too narrow on some analog channels and provoice frequencies that are being passed through to DSD+ as described further in your response.

My post previous to yours covers some strategies for tacking poor DSD+ decodes. How far away from the transmit site are you, and what antenna are you using? What system are you monitoring? Is it simulcast?

I am monitoring the Albuquerque, NM ABQ/BernCo EDACS/ProVoice system in Bernalillo County, NM. I don't know exactly where the site is, but it can't be further than 20 miles away. I assume its on top of Sandia Peak somewhere. Thats where all the towers are in this town it seems. I have no idea what simulcast means, but if it means that there are many different talkgroups on many different frequencies, simultaneously, than yes, it probably is As far as antenna. It's the junk one that came with the SDR. Very thin wire and a SMA connector. The reviews for the SDR are great except the antenna. I don't know enough to make that judgment. I'm thinking of making this one:

https://www.youtube.com/watch?v=25hvrdmzMak

You can also test through SDR# by piping the audio to DSD+ and staying on one of the voice channels (be sure to turn off all of the audio filters); however, unless your system is busy with a lot of ProVoice traffic, it could take some time for something to roll around. Regarding the bandwidth, I usually adjust mine down to a little outside of the signal on the waterfall view. 12.5kHz, which is the narrow banding bandwidth is cutting it pretty close. If your not dead center, you may be missing stuff. I'd widen it. ~20000Hz is usually what I use.

I tried this as well as mentioned earlier. In SDR# the most the bandwidth widens to is 32mhz. I tried in unitrunker to set it to 32mhz and it didn't seem to make too much difference.

Start by testing gain setting in SDR+ (where you can visualize what's going on). Then adjust the UniTrunker gain appropriately (UniTrunker gain is ~ 10x the SDR+ gain). IN SDR+, what is the best signal peak to noise floor you can achieve? Also, use SDR# to figure out best antenna placement and the best antenna you have available to you.

Once you go through the basics of achieving the best signal reception possible, there are a few tweaks in DSD+ to fine tune the decode. Your biggest improvement, however, will be in the quality of signal you provide DSD+.

You lost me a little bit here. Do yo mean DSD+ or SDR# in the first part? I also don't understand too well what you mean by signal peak to noise floor. That's in the waterfall?

Thanks for your reply man. I really appreciate it and Ill definitely read and reread this thread. Its all great information.

 

[AggieCon]

I looked up the site you are monitoring. I bet you are much closer than 20 miles. It is simulcast.

Simulcast means that multiple repeater (i.e. tower) sites work together on the same frequencies as if they were a single tower. Instead of having each tower assigned unique frequencies and deciding which traffic to carry, all of the towers in the area act as one. What this means to scanners in a practical sense is that there is increased signal distortion due to multipath, which can make digital traffic much harder to decode. It also has permutations on what type of antenna works best.

The system you are following appears that it might have a fair share of encrypted traffic. If so, that might be the source for a lot of your DSD+ problems and error codes. I don't have any encrypted ProVoice (or Voice Guard, whatever their marketing department named it) to listen to, so I can't confirm this behavior. Hopefully someone else will chime in. To test this out, in UniTrunker, lock out all of the talkgroups that have an "E" for the mode in the Radio Reference database. To lock them out, go to the system tab on the main UniTrunker window, open the correct system, go to the "groups" tab. The ~10 column over is "Lockout." Enable this for each of the encrypted talk groups.

I am not aware of a method to adjust the bandwidth in UniTrunker, so you are ahead of me there. 25kHz is probably best since a lot of stuff is till wideband, I think. It is important that your frequency correction for your RTL is accurate, otherwise, you are not tuning center on the frequency. See previous posts in this thread for info re ppm correction (http://forums.radioreference.com/software-defined-radio/333044-provoice-listening.html#post2571705).

Let me know how that antenna works for you. If you send me a screenshot of your SDR# with the control channel or some voice traffic in range, it will help others understand your situation better. You might be getting plenty of signal already. In fact, attenuating some of the towers might help you out. Or a directional antenna. Or a corner or other reflector/attenuater to modify the directionality of your antenna. For what it is worth, I am in the center of a 6 tower simulcast system, and the stock RTL antennas work fine for me. To make me feel like I did something special, I set them on some foil.

Click the picture below for a full size version. Noise floor is the strength of all the non-desirable signal in the background. The peak signal strength is how well we receive the signal we are after. It is ideal to have as large of difference as possible between the noise floor and the signal of interest. Don't worry about the gain for now. As an example, in the picture I've attached, a gain of ~39 works well for the signal on the far right. However, if I were after the middle signals, I would bump it up some. The signal on the far right is ~10,000 times stronger than the noise floor.

 

[rctelles]

I'll have to re-read the post again AggieCon. My eyes are too tired right now. I'll get some pictures sometime this weekend. Sorry John1984 for hi-jacking your thread a little bit but it's exactly what I was after and at the top of the forum so I didn't want to create another post on ProVoice.

 

[rctelles]

I looked up the site you are monitoring. I bet you are much closer than 20 miles. It is simulcast.
The system you are following appears that it might have a fair share of encrypted traffic. If so, that might be the source for a lot of your DSD+ problems and error codes. I don't have any encrypted ProVoice (or Voice Guard, whatever their marketing department named it) to listen to, so I can't confirm this behavior. Hopefully someone else will chime in. To test this out, in UniTrunker, lock out all of the talkgroups that have an "E" for the mode in the Radio Reference database. To lock them out, go to the system tab on the main UniTrunker window, open the correct system, go to the "groups" tab. The ~10 column over is "Lockout." Enable this for each of the encrypted talk groups.

Ahead of you on this. I've tried to lockout all the encrypted channels that are published. I've locked out a bunch of stuff except dispatch and a few other unencrypted channels i know i want to monitor.

I am not aware of a method to adjust the bandwidth in UniTrunker, so you are ahead of me there. 25kHz is probably best since a lot of stuff is till wideband, I think. It is important that your frequency correction for your RTL is accurate, otherwise, you are not tuning center on the frequency. See previous posts in this thread for info re ppm correction (http://forums.radioreference.com/software-defined-radio/333044-provoice-listening.html#post2571705).

On the main unitrunker page click systems. Click the system, then the channel tab. Bandwidth is an editable box. Dont know if it does anything haha. I might be off on the ppm correction. Its very close, but any tampering from 5 and 75 on each respective receiver seems to yield no change.

Let me know how that antenna works for you. If you send me a screenshot of your SDR# with the control channel or some voice traffic in range, it will help others understand your situation better. You might be getting plenty of signal already. In fact, attenuating some of the towers might help you out. Or a directional antenna. Or a corner or other reflector/attenuater to modify the directionality of your antenna. For what it is worth, I am in the center of a 6 tower simulcast system, and the stock RTL antennas work fine for me. To make me feel like I did something special, I set them on some foil.

Click the picture below for a full size version. Noise floor is the strength of all the non-desirable signal in the background. The peak signal strength is how well we receive the signal we are after. It is ideal to have as large of difference as possible between the noise floor and the signal of interest. Don't worry about the gain for now. As an example, in the picture I've attached, a gain of ~39 works well for the signal on the far right. However, if I were after the middle signals, I would bump it up some. The signal on the far right is ~10,000 times stronger than the noise floor.

Alot of this seems like greek to me, but I follow a little. I got the screenshots after careful positioning of the antenna I feel gets the best signal until I get my homebrew made. Its actually a wifi boosting 2.5ghz antenna. Dont know if that matters, but it works, and its all i have.

This is the control channel and random traffic in SDRsharp.




This is in unitrunker and DSD+ on the info. Top receiver is control, bottom is voice. The no sync is usually at the end of calls only.



This is the VCO in the same.



And thats it. The antenna is not in a permanent or probably optimum place I am sure. Hell, its not even the right antenna! Thanks for all the insight AggieCon. I appreciate it very much.

 

[AggieCon]

It looks like you are really getting the hang of this.

Here are some tips:

• On the Signal VCO, it's good that you have it muted. You can do ahead and set the Audio Output to unspecified to make sure you don't accidentally pass it through.
• On the Voice VCO, under the "voice" settings, go ahead and select "analog." Set the "Audio Out" to your speakers. It appears there might be analog traffic on the system you monitor. Right now, you would not catch any of it. On the far right of the call history window, under "Svc" there is a column that identifies the traffic as analog or digital.
• The PPM correction is very important. Did you determine the two values, 0 and 75, in SDR#? In the sceenshot of SDR# you sent, you appear to be perfectly centered, but I can't see the frequency input. Is the frequency listed at the top left the exact control channel?
• Add a squelch value of ~53 to both the Voice and Signal VCOs on the VCO tab. This will allow the drift correct function to work. It will also prevent noise from the Voice VCO.
• To reduce computing demands, you could lower the Sample Rates of your receivers since you have just 1 VCO on each. That said, sometimes it seems to work better with the higher bandwidth.
• Start adding in the "encrypted" talk groups when you want. The RR database isn't always perfect. Those groups could carry mostly clear traffic, or they might be 100% concealed. You might be able to find more information in the state forum.

It looks like you are getting signal that is more than adequate. Your UniTrunker decode looks good based on the color signal bar (I can't see the health readout... it's further down in the Signal VCO page). Your DSD+ decode is almost perfect. If you don't like how the voice sounds, you might go through the DSD+ fine tuning steps. Just keep in mind the benefit might be negligible. The "no sync" at the end is usual behavior. It's no longer in sync with the data frames when there is no more traffic coming across the radio.



How does the audio sound? It looks like you have a good setup. I hope you're happy with it. Good job!


I recommend ScannerBox.us - Streaming live audio feeds to Logitech Squeezebox Radio for your next scanning project. You can buy Virtual Audio Cable and add a couple more Voice VCOs to increase the amount of traffic you catch.

 

[rctelles]

It looks like you are really getting the hang of this.

Here are some tips:

• On the Signal VCO, it's good that you have it muted. You can do ahead and set the Audio Output to unspecified to make sure you don't accidentally pass it through.
• On the Voice VCO, under the "voice" settings, go ahead and select "analog." Set the "Audio Out" to your speakers. It appears there might be analog traffic on the system you monitor. Right now, you would not catch any of it. On the far right of the call history window, under "Svc" there is a column that identifies the traffic as analog or digital.
• The PPM correction is very important. Did you determine the two values, 0 and 75, in SDR#? In the sceenshot of SDR# you sent, you appear to be perfectly centered, but I can't see the frequency input. Is the frequency listed at the top left the exact control channel?
• Add a squelch value of ~53 to both the Voice and Signal VCOs on the VCO tab. This will allow the drift correct function to work. It will also prevent noise from the Voice VCO.
• To reduce computing demands, you could lower the Sample Rates of your receivers since you have just 1 VCO on each. That said, sometimes it seems to work better with the higher bandwidth.
• Start adding in the "encrypted" talk groups when you want. The RR database isn't always perfect. Those groups could carry mostly clear traffic, or they might be 100% concealed. You might be able to find more information in the state forum.

It looks like you are getting signal that is more than adequate. Your UniTrunker decode looks good based on the color signal bar (I can't see the health readout... it's further down in the Signal VCO page). Your DSD+ decode is almost perfect. If you don't like how the voice sounds, you might go through the DSD+ fine tuning steps. Just keep in mind the benefit might be negligible. The "no sync" at the end is usual behavior. It's no longer in sync with the data frames when there is no more traffic coming across the radio.



How does the audio sound? It looks like you have a good setup. I hope you're happy with it. Good job!


I recommend ScannerBox.us - Streaming live audio feeds to Logitech Squeezebox Radio for your next scanning project. You can buy Virtual Audio Cable and add a couple more Voice VCOs to increase the amount of traffic you catch.

All great tips. This is with the antenna I built. I did it without the pizza pan, so it just looks like a long piece of pvc pipe. Works good inside my garage and house.



I also followed the instructions for dsdtune and am stating dsdplus with "dsdplus.exe -fp -dr3 -dh7" as recommended by that program and its seemed to cut down the errors a lot. Sounds pretty good. Not as good as the web stream on broadcastify but pretty close.

I also have the virtual cable set to 33% on windows volume options. Any opinion on the paid vs free virtual cable? I will probably look into that link in your last post, because it looks like you can use it to stream. I have no intention of streaming to the net, but to my android or laptop would be awesome.

 

[AggieCon]

Your new antenna is a substantial improvement over the previous one you were using. I prefer Virtual Audio Cable. I think it is better. You'll need it if you are going to go with more than 1 Voice VCO, anyways. Trunking recorder lets you listen or "stream" through your web browser. It's awesome. If you open up the port on your router, you can listen away from home as well.

For what it is worth, I have my VAC from UniTrunker to DSD+ pretty low on volume too. I am surprised you're not getting perfect decode. You have a strong signal. Is your processor up to the task? You're running your power setting on "maximum performance," correct? See if you can contact the feed provider and see what he does. Double check your UniTrunker gain settings with your new antenna. Good job. I think you're a SDR expert now.

 

[rctelles]

Your new antenna is a substantial improvement over the previous one you were using. I prefer Virtual Audio Cable. I think it is better. You'll need it if you are going to go with more than 1 Voice VCO, anyways. Trunking recorder lets you listen or "stream" through your web browser. It's awesome. If you open up the port on your router, you can listen away from home as well.

For what it is worth, I have my VAC from UniTrunker to DSD+ pretty low on volume too. I am surprised you're not getting perfect decode. You have a strong signal. Is your processor up to the task? You're running your power setting on "maximum performance," correct? See if you can contact the feed provider and see what he does. Double check your UniTrunker gain settings with your new antenna. Good job. I think you're a SDR expert now.

Im running an overclocked i5 on my desktop and a much slower mobile celeron on a laptop. The desktop is a little better at it obviously. I am still using auto gain in unitrunker. I tried with different settings for manual gain with no difference or much luck. You lost me at power settings. Is that in unitrunker?

 

[AggieCon]

No, on your desktop or laptop in Windows, in the power settings. They ship most computers in EPA environmentalist mode. Make sure your computer isn't being throttled down.

An i5 should be enough for one instance of DSD+.

 

[rctelles]

No, on your desktop or laptop in Windows, in the power settings. They ship most computers in EPA environmentalist mode. Make sure your computer isn't being throttled down.

An i5 should be enough for one instance of DSD+.

Got you. I dont think its throttled down but wouldnt hurt to check. I also emailed the feed provider from broadcastify. They used to use a similar setup for this feed before the bc provoice upgrades so maybe they can chime in. I sent a join request to the group. Thanks for the invite!

 

[AggieCon]

I see that the feed provider is using the 436. I guess it must do a better job. I suppose I should get one. But for a feed or computer use, I think multiple voice VCO and trunking recorder is the best way to go.

 

[rctelles]

Its out of my price range right now, hence the sdr, and even that is draining my wallet haha. I contacted the feed provider for my area and he stated that provoice is not great to begin with so even a hardware based setup would give some decoding or audibility problems. That aside he stressed ppm correction as everyone else has and using the newer TCXO R820T2 dongles (I ordered two).

He also recommended to keep experimenting with the line in audio volumes and just running DSD+ stock with no tuning or switches.

I ran each dongle in SDR# for 20 minutes then noted the ppm again and I was maybe off by about 1-2 on each dongle, so Im using those new settings in unitrunker. I think this is as good as it may get. I recorded a small sample of what I am getting and clipped out some of the radio silence to make the file shorter. Here is the result:

https://www.dropbox.com/s/090qh7lp08exwth/Provoice.wav?dl=0

 

[AggieCon]

That sounds pretty good for ProVoice. One problem, which I think occurred about midway, is on the user side: Many users talk too loud into the radio / the ADC gain on the radios is too hot. ProVoice doesn't like that. When I hear traffic having that problem, parts of the correspondence is unintelligible as if the volume over the clipping point is cutoff (but it doesn't actually clip in the playback) and so whatever they were saying is just a blend of loud. You know they are saying something, but you can't differentiate it. I confirmed this by comparing the P25 to ProVoice patched traffic (on OP25, the hot gain will actually clip).

Sounds like the feed provider passed along some good advice. I mostly agree with leaving the tuning off. I experience too much variability between calls for the fine tune to be worthwhile. If you had the source audio recorded and there was something you really wanted to try to decode better, the fine tune would be good for that purpose. However, leaving it set all the time will help some of the traffic decode better, but the traffic not matching the configuration will probably decode worse.

DSD+ plus is pretty sensitive. Having too hot of source audio is worse than "too quiet".

If you have drift correct and squelch set in UniTrunker, it will handle getting to a near perfect drift correction. You just need to get it close.

I have several varieties of RTLs (except the mini), and I haven't noticed subpar performance from any of them. I have several there were less than $10.

 

[mk262]

You're best off using an airspy honestly.