Thanks for your reply.
Indeed any voltage above the schmitt trigger high input level should be interpreted as logic high. The 4 to 40V range is guranteed to be logic high.
I designed a similar circuit to your suggestion to measure frequency even in an amplitude level 50mV. But the circuit for digital input should be simpler and less complicated.
I have tried the first circuit, which is with shotky diodes, without pull down/up resistor. All the resistor were the value of 1M. It worked well even with a 30 kHz signal. But I am not sure how it behaves with a pull up/down resistor and at high temperature for the worst case. And another point to consider is how much it will degrade the performance of analog inputs when clamping occur. I have one power supply; AGND and DGND is separated with ferrite bead. AVDD and DVDD is also.
The capacitor at output of schmitt is used to remove the ringing edges. It works as expected, I have seen it on scope.
Need more suggestions...