I don't know exactly how they do it on the N54, but my understanding is that what is being listened for could be harmonics - second order and so on. That does not happen in the sensor - requires a computer algorithm to sort that out. But can certain effect the range which one would want the sensor to function.
Long ago, I understood from a engineer that was doing early Porsche development, that the reason for going after the harmonics was that the 2nd order harmonic frequency was in a far less noisy frequency band.
I don't know how one takes an existing sensor and desensitizes it without putting a filter on its output. With a filter, the only other way I would know is to literally have the sensor be a different product with a different characteristic set.
The DME reads the whole vibration bandwith the piezo element within the knock sensor is calibrated to. I would assume it is not only reading a specific harmonic but rather the DME is looking for for a specific frequency. Everything has its own resonance frequency and on top of that harmonics.
So this is the core issue of the DME. Your motor mounts, dmfw, trans, sub frame.. etc..etc.. can trigger a knock as different parts can have the same frequency in different harmonics.
So our knock tables are calibrated for stock power and if you double that hell brakes loose as tons of other parts start vibrating in other frequencies than they are supposed to and can overlay with each other or.. cause false knock as the same vibration is caught as the DME thinks only can come from the combustion.
When our NVH Team analysis our compressor they are also looking for frequencies/harmonics that are suspicious. When they present the results to us it looks like this.
You have the base frequency and all (or a lot) harmonics recorded. The DME gets the whole vibration transferred to the knock sensor. The piezo element in the knock sensor gets the vibration and transforms it into an electrical signal which is then forwarded to the DME. The DME takes the knock sensor tables and pulls timing when the signal triggers "knock".
If it is the first or second order i don´t know. But there is nothing like marty describes except it would be delivered with additional electronics that filter out things before forwarded to the DME.
The VTT crap is just a stock knock sensor with a different resistor. So just a desensitized sensor so that our DME doesn´t trigger as early. The same thing as back when BMS put a resistor between the sensor cable and the dme.
The basis of the VTT "custom" sensor. No wonder they are not sold anymore.