I've been used to the pristine audio quality of KCERS and am finding PSERN to definitely be a big step backwards. I'm using RTL-SDR-based dongles with trunk-recorder and I've verified a strong signal on waterfalls and have the error Hz within +/- 100 Hz so I'm not sure that I have much room for improvement in my setup. Yet, I find calls to sound pretty terrible. Maybe I'm just too used to how the analog system sounded and my ears just some adjusting?
PSERN Sound Quality.. bad?
- Thread starter louisik1
- Start date
- Status
1: What are your gain settings? Too high? What is your noise floor?
2: What antenna(s) are you using? Are you using a multi-coupler if using more than 1 RTL-SDR?
3; "Waterfalls" - Are you using SDR-Trunk? (If not, try using it).
4: What site are you using? How far from it are you?
Bottom line: There are many variables that will affect P25 II sound quality when using an RTL-SDR set-up.
PSERN, when using a properly set up radio or RTL-SDR set-up sounds pretty good.... if.......
2: What antenna(s) are you using? Are you using a multi-coupler if using more than 1 RTL-SDR?
3; "Waterfalls" - Are you using SDR-Trunk? (If not, try using it).
4: What site are you using? How far from it are you?
Bottom line: There are many variables that will affect P25 II sound quality when using an RTL-SDR set-up.
PSERN, when using a properly set up radio or RTL-SDR set-up sounds pretty good.... if.......
Whoops, sorry about my previous response, I thought I deleted it.
But yes there is something wrong. When the PSERN system first went online I heard one of the security officers doing work at the stadium say how it sounded like people have marbles in their mouth lol. And the other guy responded yeah it sounded like that to him too.
There are certain elements and broadcasts that sound perfect. Usually the dispatchers sound fine. And when they're holding the air and using a tone, the tone sounds perfect. But some of the responding police sound like they have a super flange going on. And a lot of those times, when it's really bad, the dispatcher will ask the officer to repeat themselves. That is a relief to me because it tells me that it's not just my setup. And like the poster before me, my setup is strong, 100% lock with a deviation of less than 10.
BUT... When I have used my bearcat scanner, it sounds better. I only did a quick test back when the system first switched over, but I haven't done a comparison since. I'll try to get it out and get it programmed this weekend and see what I come up with.
If you want to listen to one of my current PSERN feeds, listen to Seattle Police and Fire. It should become apparent quickly.
But yes there is something wrong. When the PSERN system first went online I heard one of the security officers doing work at the stadium say how it sounded like people have marbles in their mouth lol. And the other guy responded yeah it sounded like that to him too.
There are certain elements and broadcasts that sound perfect. Usually the dispatchers sound fine. And when they're holding the air and using a tone, the tone sounds perfect. But some of the responding police sound like they have a super flange going on. And a lot of those times, when it's really bad, the dispatcher will ask the officer to repeat themselves. That is a relief to me because it tells me that it's not just my setup. And like the poster before me, my setup is strong, 100% lock with a deviation of less than 10.
BUT... When I have used my bearcat scanner, it sounds better. I only did a quick test back when the system first switched over, but I haven't done a comparison since. I'll try to get it out and get it programmed this weekend and see what I come up with.
If you want to listen to one of my current PSERN feeds, listen to Seattle Police and Fire. It should become apparent quickly.
Hey, thanks for the replies. 
Here’s more about my setup:
Remtronix 820S antenna -> coupler -> FM notch filter -> LNA (+20dB) -> 2x Nooelec RTL-SDR dongles. They are hosted by a Raspberry Pi using rtl_tcp. I have a relatively short USB cable going to my Raspberry Pi with ferrite chokes on each end. I have a PC running Linux using trunk-recorder to handle the broadcasting duties as the Pi couldn't keep up with the amount of recorders necessary to capture the huge amount of calls that go over PSERN. trunk-recorder uses osmosdr and gnuradio as its stack.
Since I use rtl_tcp, it means I can't use SDRTrunk since it doesn't support devices that aren't physically connected. Once I get a longer cable for the USB hub that has the dongles connected to it, I can try that and see if I get any better results. (I don't need a super long cable -- 6' should suffice, but my USB hub requires an oddball USB-A to USB-A cable and that's the one kind of cable I don't have in bulk.)
In my trunk-recorder settings, I've been experimenting with different gain settings on the SDR itself. I've found 16-20 seems to be a good value but I'm still tweaking things. My RTL-SDRs have a good 0.5PPM TCXO and I'm observing minimal frequency drift after tuning the PPM values in my trunk-recorder config.
I'm not sure what my noise floor is or how best to check it but I'm happy to dig into this more. I'm assuming it's not great. My antenna has to be indoors for various reasons but it's somewhat isolated from direct noise sources.
RRDB doesn't seem to have coordinates for PSERN towers so I'm not sure how close it is. When I was broadcasting KCERS, I was using the Education Hill tower which is line of sight for me. If PSERN is using the same infrastructure, then it's definitely close.
Here’s my feed if anybody wants to listen and tell me if it really sounds as bad as I think it does: OpenMHz.
Here’s more about my setup:Remtronix 820S antenna -> coupler -> FM notch filter -> LNA (+20dB) -> 2x Nooelec RTL-SDR dongles. They are hosted by a Raspberry Pi using rtl_tcp. I have a relatively short USB cable going to my Raspberry Pi with ferrite chokes on each end. I have a PC running Linux using trunk-recorder to handle the broadcasting duties as the Pi couldn't keep up with the amount of recorders necessary to capture the huge amount of calls that go over PSERN. trunk-recorder uses osmosdr and gnuradio as its stack.
Since I use rtl_tcp, it means I can't use SDRTrunk since it doesn't support devices that aren't physically connected
. Once I get a longer cable for the USB hub that has the dongles connected to it, I can try that and see if I get any better results. (I don't need a super long cable -- 6' should suffice, but my USB hub requires an oddball USB-A to USB-A cable and that's the one kind of cable I don't have in bulk.)In my trunk-recorder settings, I've been experimenting with different gain settings on the SDR itself. I've found 16-20 seems to be a good value but I'm still tweaking things. My RTL-SDRs have a good 0.5PPM TCXO and I'm observing minimal frequency drift after tuning the PPM values in my trunk-recorder config.
I'm not sure what my noise floor is or how best to check it but I'm happy to dig into this more. I'm assuming it's not great. My antenna has to be indoors for various reasons but it's somewhat isolated from direct noise sources.
RRDB doesn't seem to have coordinates for PSERN towers so I'm not sure how close it is. When I was broadcasting KCERS, I was using the Education Hill tower which is line of sight for me. If PSERN is using the same infrastructure, then it's definitely close
.Here’s my feed if anybody wants to listen and tell me if it really sounds as bad as I think it does: OpenMHz.
Try it without that LNA...
You may have a strong enough signal without it.
Also, the LNA is probably hiking up the noise floor quite a bit.
Which certainly could add to P25II decoding quality problems.
BTW: myself (and a few others here) have the actual PSERN individual tower locations.
Someone should be able to help determine which site is closest to you.
I'm going to be off-line for a few days, so if you still need help, reach out later in the week....
You may have a strong enough signal without it.
Also, the LNA is probably hiking up the noise floor quite a bit.
Which certainly could add to P25II decoding quality problems.
BTW: myself (and a few others here) have the actual PSERN individual tower locations.
Someone should be able to help determine which site is closest to you.
I'm going to be off-line for a few days, so if you still need help, reach out later in the week....
freqseeker
Member
I listened for a few minutes and I think it sounds fine. I didn't hear any break ups or signal issues, but that can vary by time of day.
I've only started the PSERN listening the past 48 hrs but it sounds fine by me. I'm using SDS100 and SDS200 with factory antenna and am located just N or Boeing Field on the top of Beacon Hill (near VA hospital). It's obvious that they're digital though vs analog but it's no different than analog telephone vs digital
I've verified more TGIDs for Seattle PD and FD but not sure how to incorporate and update them into my database yet. That'll be homework for me this week (moved up to SDS radios less than a year after my RS Pro2006 for a couple dozen years). Much to learn
I've verified more TGIDs for Seattle PD and FD but not sure how to incorporate and update them into my database yet. That'll be homework for me this week (moved up to SDS radios less than a year after my RS Pro2006 for a couple dozen years). Much to learn
I set up my Uniden BCD9962P2 and am listening to PSERN now. It sounds the same in terms of flanging, distortion, and 'marble mouth' as it does from the SDRs and my broadcasts. It isn't every broadcast/dispatch, but its enough, maybe 1/3rd of everything heard, to recognize there's a problem. I'm surprised it isn't more obvious to other people. Another way to describe it (and maybe closer to the real issue), is it sounds like low quality mp3s, like vbr 8khz or something. Blech. I guess the upside for me is the confirmation that my setup is fine.
freqseeker
Member
I agree with you on the quality being bad. It does sound typical to me though.
That's how I feel. We learn to accept shortcomings for convenience. There's very little quality anymore.
I guess I would feel less concerned about the sound quality if people's lives weren't at stake.
true...but I'm willing to bet that their equipment is much more refined and they have quality audio tx/rx. I'll try to get arrested or something and get inside. Seriously, a couple relatives are LEO so I'll see if I can ask them or ride along
Even in the analog days, a professional public safety radio will beat a scanner. It's why they cost so much more, so don't blame the system.
Believe me, they've done hours and hours of range testing and other performance checks before it's released
Believe me, they've done hours and hours of range testing and other performance checks before it's released
kingpin
Trailer Park Supervisor
It's honestly not really a fair comparison in judging the audio quality of a $20 SDR and free software against radios and equipment that costs thousands, and is task made with better processing, filters, and robust audio components. I promise you, the audio sounds great on the system radios. There is always a break-in period for your ears when going from analog to FDMA and even TDMA. It's gonna sound different, but the difference is less noticeable for the end users the system is designed for. It's not built to entertai scanner land. You'll get used to it
devicelab
Whacker Extraordinaire
The SDS-100 has slightly better audio decode than the APX. It's definitely more consistent in the digital quality. The SDS lacks the punch of the Motorola but you can sort of compensate for that with higher volume.
I haven't confirmed this but I believe each TGID on the system has its own settings and individual radios may need tweaking. SPD talkgroups (generally) sound excellent. Other agencies are hit or miss. I wouldn't think there's a large (significant) difference between the APX models but there could be...
- Status
Similar threads
