Study Of Frequency Spectrum Refinement Algorithm Based On Fast Fourier Transform

With the continuous development of society, Science and technology and computerapplications continuously improve, people want to analyze things more quickly andaccurately. For example, machine tools, ships, or large-scale machinery and equipmentby computer monitoring to fault diagnosis; using sonar system to analyze the size,location of unidentified object underwater. If you want to analyze the state of thesethings, you must precisely and rapidly analysis the signal of these things, spectrumanalysis is the rise in the conditions. The detected signal contains information mixedwith some useless information, it will affect the final analysis results, getting the wrongjudgment. So spectrum refinement method becomes Spectral analysis of the mostpowerful tool, By improving the frequency resolution, displaying all information,extracting useful information for analysis, finally getting the correct judgment.Spectrum refinement method does not increase the sampling points and reduce thesampling frequency, using other ways to improve the frequency resolution. This paperanalyzes several spectrum refinement methods, this paper presents a spectrum zoombased on two-section FFT algorithm, the simulation results show the feasibility of thenew method.The main contents of this paper have the following aspects:(1) The Fourier Transform is a basic method of spectral analysis, and is also a veryimportant one, which makes the signal analysis and processing theory truly can beapplied in practice. This paper analyzes the spectrum of the two different types ofrefinement method, simultaneous analysis of their advantages and disadvantages.(2) This paper describes the basic principles of the Fast Fourier transform, highlights theFFT and CZT spectrum of joint algorithm refinement algorithm, by analyzing theirrespective principles and advantages. Through joint two algorithms, makes thefrequency spectrum analysis become higher resolution and higher precision.(3) In this paper, the research on the mobile FFT algorithm, presents a spectrum zoombased on two-section FFT algorithm. By increasing its mobile data length, makesthe length of the mobile data to N/2, then the use of storage space in exchange forthe calculation time, reducing the amount of computation of a single, although thetotal calculation amount is not changed, but due to increase the actual calculationdata length, so that the frequency resolution is increased and the operation speed isnot reduced. Simulation results show the feasibility of the new method theory.