Experimental version with option to disable CC filtering

VK3KYY
Posts: 7478
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental version with option to disable CC filtering

Post by VK3KYY » Sun Feb 23, 2020 3:25 am

Ik0nwg wrote:
Sun Feb 23, 2020 2:56 am
VK3KYY wrote:
Sat Feb 22, 2020 5:25 am
Doing some more testing, I think TA is not always working if the CC does not match.

I'm not sure why this is happening at the moment.

Edit.
It looks like the ID is OK, but its the TA that is not being decoded.
ciao Roger
drove around with the vfo scan activated and the CC filter disabled for several hours, everything seems to work fairly well:
- small noises
- on low signals there is some uncertainty in decoding and the audio is not always perfect.
But his work of decoding all CCs does it and therefore goal achieved.
You're right it is only the TA that is not displayed, id dmr appears (sometimes late) but is displayed. It would be useful to view the decoded information (CC number) but obviously this is not a priority. I read that you are working to solve these small problems and thank you.
Sal
Colin has sent me an update to improve the audio etc, but I'm not sure if it helps the ID display.

I've been analysing some DMR traffic, and I think we can generally improve the acceptance criteria, based on the various attributes of the DMR signal data, to make this even better.

Currently we are mostly still using the acceptance criteria developed by Kai, but I think we may be able to improve things a little

NA7Q
Posts: 151
Joined: Wed Jan 01, 2020 3:41 pm

Re: Experimental version with option to disable CC filtering

Post by NA7Q » Sun Feb 23, 2020 3:33 am

Bow howdy, some of the horrendous noise I heard on this was bad. I was listening to my hotspot with my locals talking, and I decided to switch between each of the settings to see what happened. A couple different times as I switched between each filtering level a noise burst out of nowhere, changing the setting did not fix it. I had to reboot. So far I haven't replicated it to that extent again.

VK3KYY
Posts: 7478
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental version with option to disable CC filtering

Post by VK3KYY » Sun Feb 23, 2020 3:38 am

NA7Q wrote:
Sun Feb 23, 2020 3:33 am
Bow howdy, some of the horrendous noise I heard on this was bad. I was listening to my hotspot with my locals talking, and I decided to switch between each of the settings to see what happened. A couple different times as I switched between each filtering level a noise burst out of nowhere, changing the setting did not fix it. I had to reboot. So far I haven't replicated it to that extent again.
Thats why its called experimental ;-)

VK3KYY
Posts: 7478
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental version with option to disable CC filtering

Post by VK3KYY » Sun Feb 23, 2020 5:52 am

Colin sent me an update to the code, which slightly improves the noises at the end of the Rx signal.

It doesn't however resolve the problem with the ID not displaying, and I need to do a complete review of the various acceptance criteria algorithms that are currently in the firmware.

All of the acceptance code was written by Kai, about 9 months ago, by I think reverse engineering the criteria based on the DMR data which he captured.

However to handle the CC not being filtered, I think things need to be changed.
Attachments

[The extension sgl has been deactivated and can no longer be displayed.]


TA2ISO
Posts: 19
Joined: Sun Feb 16, 2020 1:41 pm
Location: Istanbul

Re: Experimental version with option to disable CC filtering

Post by TA2ISO » Sun Feb 23, 2020 3:10 pm

I just tried the latest version and have no garbled noise anymore.

User avatar
KV0A
Posts: 16
Joined: Tue Dec 17, 2019 11:19 pm

Re: Experimental version with option to disable CC filtering

Post by KV0A » Sun Feb 23, 2020 4:45 pm

VK3KYY wrote:
Sun Feb 23, 2020 3:23 am
KV0A wrote:
Sun Feb 23, 2020 2:43 am
I'm still a little unsure how the Ct filtering works. Does this make it so that if you're on a private call then everything except the private call is filtered?
Colour Code on DMR is a bit like CTCSS on FM.

It does not make the call private, but if you radio is not set to the same Colour Code as the other station, the radio may not even put on the green LED to indicate that there is a signal at all.

Most Ham radio stations and repeaters seem to use CC 1, however there are some repeaters which use other CC numbers - and this has confused quite a few people, who presumed that everything was on CC1.


Apparently repeaters in close proximity to each other which are on the same frequency, are supposed to use different CC, but I'm a bit sceptical about how effective this would actually be.

CC has a special use for DCS but I don't know the technicalities of this.
I'm familiar with the color code (though I'm unsure if it's just bits in the packet header or if it's more like CDMA where different symbols are used based on the CC), and my understanding is that it's supposed to prevent repeaters on the same freq from conflicting with each other.

But I was actually asking about the Contact (Ct) filtering. If you turn that on then does it filter out everyone who is not a "private call" contact in the radio? Or does it just silence everything but an incoming private call?

NA7Q
Posts: 151
Joined: Wed Jan 01, 2020 3:41 pm

Re: Experimental version with option to disable CC filtering

Post by NA7Q » Sun Feb 23, 2020 5:39 pm

TA2ISO wrote:
Sun Feb 23, 2020 3:10 pm
I just tried the latest version and have no garbled noise anymore.
I just tested with my radio, and holy cow. It's still pretty horrendous using my hotspot. Very experimental.

G4EML
Posts: 919
Joined: Sat Nov 16, 2019 10:01 am

Re: Experimental version with option to disable CC filtering

Post by G4EML » Sun Feb 23, 2020 5:52 pm

I think that goes to show how poor the signals from some hotspots really are. They rely pretty heavily on the radios to clean things up.
Disabling colour code detection is not really relevant to hotspot use anyway as you are always going to know what it should be set to.

TA2ISO
Posts: 19
Joined: Sun Feb 16, 2020 1:41 pm
Location: Istanbul

Re: Experimental version with option to disable CC filtering

Post by TA2ISO » Sun Feb 23, 2020 10:29 pm

NA7Q wrote:
Sun Feb 23, 2020 5:39 pm
TA2ISO wrote:
Sun Feb 23, 2020 3:10 pm
I just tried the latest version and have no garbled noise anymore.
I just tested with my radio, and holy cow. It's still pretty horrendous using my hotspot. Very experimental.
I don`t use hotspot for testing, the latest experimental firmware works just as before for me. I had my HT monitor private and group calls and there are no weird noises anymore.

VK3KYY
Posts: 7478
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental version with option to disable CC filtering

Post by VK3KYY » Sun Feb 23, 2020 11:33 pm

I think we will be able to fix the strange noises, which are caused by the hotspots sending random data after the end of the real transmission, however it will require a lot of modifications to the code

Post Reply