Analysis of Speech Signals based on Short-time Fourier Method
Download as PDF
DOI: 10.23977/icamcs.2018.076
Corresponding Author
Yu Jie
ABSTRACT
Aiming at the problem that the heterodyne signals inherent in pseudo-random (PN) code modulated speech signals cannot be equally spaced, a new short-time Fourier transform (DFT) method for non-uniform sampling signals is proposed. Firstly, the speech signal system model of distance velocity measurement is given, and the reason why the heterodyne signal can’t be equally spaced is pointed out. Then, a new spectrum analysis method of non-uniform sampling signal is proposed by theoretical derivation. Finally, the simulation is verified by simulation. The method can be used to analyze the spectrum of non-equally spaced sampled signals. The results show that the method can still effectively analyze the frequency of unequal interval sampling Doppler signals in the Doppler frequency range generated by road moving targets with the received signal signal-to-noise ratio (SNR) of 0 dB.
KEYWORDS
Non-equal interval sampling, fast Fourier transform, signal processing, spectrum analysis, speech signal, road target perception