But I don’t garantee that I can answer all questions :o) I am not (yet?) an expert.
Anyway, for RxQual, I can give a few more details here; It’s strongly related to the channel coding of the messages sent over the air interface. Some extra bits are added to the messages in order to:
1) detect errors
2) correct errors
So, if you have 1 bit error in 1 frame, no problem. The error will be corrected thanks to the redundancy introduced by the channel coding. The person on the other side of the communication will not hear anything strange as he/she will receive the complete frame with no error.
Some other techniques are used in order to make the channel more failproof, like interleaving which spreads data over different frames (because heavy corruption of bits generally happen in a bursty way, not as a continuous process). So you can tolerate some bit errors and still have a perfect communication quality.
But when you lose a complete frame, that’s another story. Some information is lost and you cannot recreate the full frame. This will impact the quality of the voice communication. So for voice quality, Frame Erasure Rate (FER) is more relevant than RxQual.