After taking a look at the CQRobot GB15979 module I can see that if you wish to dive deeper into the signal path it would appear that the LM358 dual OpAmp uses the first OpAmp to condition the signal to an output voltage and applies that to OpAmp 2 at pin 6. When the voltage at pin 6 drops below the set value at Pin 5 the output at pin 7 goes low and the LED is illuminated.
To dive deeper into the signal coming from the radar sensor a connection to pin 2 could be applied to the A/D pin of the ESP or Arduino and this could provide a less binary output. Perhaps this would allows the software to be able to sense more detail from the units signal.
Of course at 10Ghz things are happening pretty fast but the considerable R/C components on the board may limit the speed that one can actually see the actual RF signal change.