Amplitude spectrum is one of the most commonly used mathematical quantitative analysis tools in seismic data analysis, and its theoretical basis is the Fourier transform. Through the amplitude spectrum analysis, we can convert the time domain data to the frequency domain to observe and compare the data from the two aspects of amplitude and phase. At present, according to the different calculation methods, the spectral analysis of the time domain signal after the Fourier transform mainly has different forms, such as amplitude spectrum, power spectral density, and noise density. This paper according to the corresponding calculation formula of three frequency spectrum dimensions of the form firstly, and then classify the normalization method commonly used in data analysis, and an analysis was conducted on the amplitude spectra of velocity\acceleration geophone, combination effects, and coupling effects. It is believed that when conducting spectral analysis, the dimensions of spectral analysis should be clearly defined, and then compared to reveal the physical properties of spectral analysis results, clarify the mathematical relationship between data performance and physical background, ensure data quality, and improve exploration effectiveness.