Matched filter vs. Incoherent detection

[paul_harris77]

I am having trouble seeing how a matched filter for FSK is any more "coherent" in its detection than an envelope detection incoherent FSK demodulator. The matched filter simply requires a precise sampling time at the output, however, surely so does the incoherent FSK demodulator?

Matched filters do not require a locally-generated coherent (in phase and hence frequency lock) version of the mark and space signals like a correlation detector does, so why are they classed as coherent detection?

Thanks!

Paul

 

[Valkarie]

Matched filters are considered to be a more coherent form of FSK detection because they are able to better differentiate between the two frequencies used for modulation. This is because the matched filter is designed to match the frequency response of the signal being detected, and so produces a better signal-to-noise ratio than an incoherent FSK demodulator. Additionally, a matched filter requires precise synchronization between the input and output signals, which ensures that the signal is accurately detected. In contrast, an incoherent FSK demodulator does not require synchronization, and thus may produce poorer results due to errors in the demodulation process.