Experimental version to improve invalid CC problem when CC filtering is disabled

VK3KYY
Posts: 2417
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia
Contact:

Experimental version to improve invalid CC problem when CC filtering is disabled

Post by VK3KYY » Tue Mar 03, 2020 9:39 am

I doubt this is a perfect solution, but I think I may have partially fixed the problem where the CC is set to zero (0) if you transmit on a simplex or to a repeater which does not respond.

I'm only updating checking and updating the CC when the CACH bit is set in the DMR data.

I did try using the CRC flag bit, but this didnt seem to help, however if I only analyse the CC when the CACH flag bit is set, seems to improve things a lot.

One thing I have noticed, is that if you tune into an existing signal with the TS and CC filter disabled, it sometimes takes a long time before the DMR audio starts to come out of the speaker.
The DMR TG and station callsign etc is shown, but there is no audio.

I'm not sure why this is happening, but its not top priority for me to investigate as the feature mostly works OK.
Attachments
opengd77_202003032021.sgl
(477.06 KiB) Downloaded 52 times

KC7RBW
Posts: 169
Joined: Wed Nov 20, 2019 1:23 am

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by KC7RBW » Wed Mar 04, 2020 2:03 am

I can't trigger the CC set to 0 bug on this firmware. Thanks!
Andrew | KC7RBW | ajorg

VK3KYY
Posts: 2417
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia
Contact:

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by VK3KYY » Wed Mar 04, 2020 2:20 am

KC7RBW wrote:
Wed Mar 04, 2020 2:03 am
I can't trigger the CC set to 0 bug on this firmware. Thanks!
OK.

I'll push this change into the Master version and make a new Latest version

User avatar
Ik0nwg
Posts: 96
Joined: Sat Nov 16, 2019 7:23 am
Location: JN61VG

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by Ik0nwg » Wed Mar 04, 2020 9:40 am

the CC0 problem seems solved, before when the CC filter was not active and stopped on the signal it made you feel the modulation and the CC number appeared ... but then strangely at the end it always positioned itself on the CC0. While that problem of not hearing modulation even when they are talking is still present. it is mainly highlighted when you turn on the radio and it lasts a few seconds. then everything seems to be working properly.
I hope I haven't said some nonsense.
thank you
ciao Roger
Sal Italy jn61vg

VK3KYY
Posts: 2417
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia
Contact:

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by VK3KYY » Wed Mar 04, 2020 9:52 am

Ik0nwg wrote:
Wed Mar 04, 2020 9:40 am
the CC0 problem seems solved, before when the CC filter was not active and stopped on the signal it made you feel the modulation and the CC number appeared ... but then strangely at the end it always positioned itself on the CC0. While that problem of not hearing modulation even when they are talking is still present. it is mainly highlighted when you turn on the radio and it lasts a few seconds. then everything seems to be working properly.
I hope I haven't said some nonsense.
thank you
ciao Roger
OK.

The problem of no audio is something different.

VK3KYY
Posts: 2417
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia
Contact:

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by VK3KYY » Sat Mar 07, 2020 7:01 am

I've been looking at the problems with the CC filter, and its difficult to resolve them.

The change I made assumed that the Rx signal was from a repeater, and the DMR data includes the CACH flag.

But I've now realized that this prevents it working on simplex, as the DMR radios are in Active mode in simplex and don't send the CACH flag.

I can change the code so that if the radio (Rx) is tuned to a repeater, that it uses the CACH flag, but does not use it, if the Rx and Tx freq's are the same and the receiving radio, i.e your radio is in Active mode.

But I'm not sure if this is a perfect solution.

There are 2 other annoyances with this feature.

1. The CC for the Channel or VFO is changed to whatever the detection algorithm finds, and you have to use the Channel details screen to set the CC back to whatever it was prior to it being changed by the detection algorithm

To fix this. I'm going to change the code, so that it does not update the master CC for the channel (or VFO) unless you press the Blue button while the signal on that CC is being received.
This is the same functionality as for the TG and TS functions of the Blue button when receiving a TG other than the TG assigned to the channel.

2. The current algorithm, seems to have problems if it only receives short bursts of signals, and can seem to get confused and never lock onto the CC.
So I need to find out why this is.




Also, its currently too easy to accidentally store the Channel details, including an invalid CC, simply by going into the Channel details screen and leaving it by pressing the Green button.

So I'm going to reinstate the requirement to press the Blue button at the same time as the Green button to make the change to the Channel permanent by writing it back to the codeplug memory

KC7RBW
Posts: 169
Joined: Wed Nov 20, 2019 1:23 am

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by KC7RBW » Sun Mar 08, 2020 1:45 am

I mentioned this on a PR but I'll be quite happy with both of these changes.

I forget if there was a way it will indicate that the settings aren't saved (or are temporary?) when they aren't.
Andrew | KC7RBW | ajorg

VK3KYY
Posts: 2417
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia
Contact:

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by VK3KYY » Sun Mar 08, 2020 2:33 am

KC7RBW wrote:
Sun Mar 08, 2020 1:45 am
I mentioned this on a PR but I'll be quite happy with both of these changes.

I forget if there was a way it will indicate that the settings aren't saved (or are temporary?) when they aren't.
I had a go yesterday at improving the CC detection algorithm by using a combination of Colin's and my methods, but could not get it to work, but that was probably as I was trying to look at it at 10pm when I need a clear head to do that sort of thing.

I've already pushed an update so that the Channel details screen does not write back to the codeplug unless Function + Green is pressed.

And I started to look at a way to indicate that the received CC was not the current CC and how to handle that.

Normally we display the data in inverse video if its not the Tx value e.g. Rx of another TG. But with the CC filter we are using the inverse video to indicate that CC filtering is disabled.

Perhaps we should just turn the "C" into inverse video when the filtering is disabled, and then change the number to inverse video if its not the same as the CC assigned for the channel.

I also experimented with no immediately updating the Tx CC, and think this is better, than what we have at the moment, but I need to add code, so that pressing the Blue button whilst the signal is being received, causes the Channel or VFO to get updated with the CC value.

However, perhaps I should not write this value back to the codeplug for the Channel, as this could be a major cause of problems

G4EML
Posts: 218
Joined: Sat Nov 16, 2019 10:01 am

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by G4EML » Sun Mar 08, 2020 11:05 am

I don’t think there is any need to change the way the indication works. If you are set to filter none changing the TX CC to be the received value makes sense and allows you to reply to a call immediately. As it’s not saved to the codeplug unless you make a deliberate action it’s only temporary.

G0NHF
Posts: 40
Joined: Sun Jan 19, 2020 10:28 am

Re: Experimental version to improve invalid CC problem when CC filtering is disabled

Post by G0NHF » Sun Mar 08, 2020 12:06 pm

Roger, did you get my private message on this subject yesterday?
Tony
Last edited by G0NHF on Sat Mar 14, 2020 5:21 pm, edited 1 time in total.

Post Reply