frequency analysis

Student ID19011210599

Analysis method of embedded cattle reading guide spectrogram

Logarithmic transformation of embedded bovine nose spectrum time domain FFT2 function.

What information can be seen from the spectrum?

The frequency spectrum embedded in cow text is a frequency domain image obtained by frequency domain transformation of time domain signal.

? 1, spectral classification

? Linear amplitude spectrum

? Logarithmic amplitude spectrum: 20log(A) transform makes the small component of the original A higher than the large component of A, thus observing the periodic signal hidden in low amplitude noise.

? Self-power spectrum: firstly, carry out correlation convolution on the measured signal to remove random interference and noise, keep and highlight the periodic signal, and lose the phase characteristics before Fourier transform.

2、? Spectrum analysis: The spectrum of the signal tells us which sine and cosine functions and frequency components the signal contains. Generally, we are concerned with the amplitude of the signal spectrum. The degree of spectrum, phase noise, frequency and stability can also be reflected in the time domain. If the time domain waveform changes sharply, there are more high-frequency components, and if the time domain changes slowly, there are more low-frequency components. The amplitude represents the energy value of the signal at this frequency. Spectrum is a complex number, and its absolute value is generally taken for the convenience of analysis. In fact, there is no imaginary number in the activity, just because it is found in the process of signal analysis that some signals can be completely divided into in-phase and orthogonal components, so setting imaginary numbers is convenient for signal analysis)

3.fft2 () function: The Fourier transform obtained by this function is near the low frequency. In order to get a better intuitive feeling, people often use fftshift () to adjust the results, so that the low frequency is in the middle. Central, changed the frequency distribution.

4. Logarithmic transformation: Because the dynamic range of spectrum is too large and the display range of oscilloscope is limited, this transformation is adopted to reduce the dynamic range of spectrum.