Optimum pre-emphasis for FM transmission with frequency-domainvoice scrambling

摘要

Optimum pre- and de-emphasis are theoretically investigated for FMnand PM transmission, including frequency-domain scrambling for voicensecurity. Using a linear system model, the optimum frequencyncharacteristics for pre- and de-emphasis filters are given as functionsnof the voice signal spectrum, demodulator output noise spectrum, and thenscrambling method. Comparisons of the maximum attainable SNRn(signaloise ratio) are made for various cases including plain voicen(nonscrambling) and secured voice scrambled by spectrum inversion,nreverberation, and DFT (discrete Fourier transform) spectrumntransformation, assuming simple voice and noise spectra