[arm-allstar] DTMF Decoding

Glenn Lyons Glenn at lyonscomputer.com.au
Thu Jan 16 17:18:52 EST 2020


I am trying to get a better understand the many factors that affect DTMF
decoding.
Accurate DTMF decoding is essential to AllStarLink/Hamvoip functionality.

Firstly, where does the decoding actually occur? Is it a DSP routine on the
Rpi? Are there any diagnostic tools at that level?

My list of factors to investigate includes....

- Quality of the source DTMF encoding. Some radios (handhelds) work better
than others. It is usually the same character that fails to be decoded when
a HH does not work.
- Quality of the RF transmission. Good antennas and transmission lines to
product a clean strong signal.
-  Audio levels in the receiver and or it's interface board. Each radio
model has it's own specification that have to be understood and adjusted.
-  De-emphasis setting in the receiver. I have read that audio levels and
de-emphasis are key. I have heard that  "twisting"  occurs if set
incorrect. What is twisting and how do you detect it? Is a
oscilloscope helpful?
-  Audio levels in the RIM. RIM usually have pots or jumpers that have to
be set in conjunction with the radio 's output to control the signal levels
entering the RIM.
-  Audio levels set in the application software. These can be set with the
very helpful admin utility in HamVoip.
-  Processing capabilities of the CPU (Rpi). Is this a factor?

Getting a clean audio signal has usually proved easy for me, however,
getting DTMF
to decode reliable has proved more difficult and caused a great deal of
frustration.

Any insights or pointers would be greatly appreciated.

Glenn
VK4PK QTH:QG62QI
DMR-MARC ID: 5054273
AllStar ID:48253


More information about the ARM-allstar mailing list