Frequency measurement based on virtual instrument software system design – the e
Introduction Including ballistic radar speed, we must complete the projectile velocity measurements, Doppler velocimetry under the principle of the need for measuring the Doppler frequency estimation, while the frequency estimation is a classic signal parameter estimation problem. At home and abroad have made a lot of ways, the main and sometimes domain, spectral estimation method and time of a frequency domain, time domain method in which a number of cycles the main zero-crossing detection method and two, the main disadvantage is low accuracy; and Time frequency domain method of a main short time Fourier transform, Wigner distribution of a Virgin, while the calculation generally higher, difficult to complete real-time processing.
In addition, the development of virtual instruments with the aid of their good man-machine interface and powerful signal processing, signal processing platform for the construction are also becoming a trend. In essence, the virtual instrument is the instrument technology and computer technology combined with deep product, it stressed that "software is the instrument" concept that allows users to define their own needs instrument function, they need to better set up a test system . It is based on signal processing and acquisition (ADC), data analysis and processing (DSP), the results of the output (DAC) and display the structure model of a common signal processing hardware platform, in this general signal processing hardware platform, called different test software on different functions of the apparatus constitutes a [1]. Based on this, this virtual instrument platform, combined with the power spectrum estimation and the combination of measurement frequency measurement, frequency measurement system for the design.
1 Frequency Measurement Often, as a useful signal and noise spectral characteristics are different, power spectrum estimation method is a noise background in the extraction of useful signals (eg sine wave) is an effective way. In view of this, the Doppler frequency measurement, data can be collected before the power spectrum analysis, and then measuring frequency by frequency domain methods to obtain complete Doppler frequency. In the power spectrum analysis can be divided into the classical and modern spectral estimation spectral estimation, classical spectral estimation method typical of a periodogram, Welch law; and modern spectral estimation methods are typical of AR model method, MA modeling, ARMA model method, entropy method, maximum likelihood and characteristics of decomposition method. Frequency Domain Frequency Measurement There are energy center of gravity method, peak search procedure, by choosing a different power spectrum analysis and frequency domain methods of measuring frequency combination, can achieve the purpose of measuring signal frequency.
1.1 Welch power spectrum estimation
Towel in the classical power spectrum estimation, using direct periodogram estimate of spectral properties is often not good, mainly relatively large fluctuations in the spectrum, the variance is relatively large. Welch method can be improved by direct periodogram estimate the variance of spectral features. The basic idea is to use sub-windowing method to a data length N XN (n) into L segments, length of M, and allow each section of data, some overlapping, were calculated for each section of the power spectrum Pi ( ), then average them to obtain the average power spectrum after , Namely:
Known under the theory of probability and statistics: Using the Welch method to estimate the variance of power spectrum is roughly a direct estimate of the spectrum of the variance of 1 / L, and the section the more the smaller variance, but the frequency resolution the lower the deviation becomes larger. So, in actual use should take into account the requirements of the variance and resolution selection of appropriate values of L and M [2].
1.2 AR model spectrum estimation method
AR model spectrum estimation method to do the principle is: the analysis assumes that the signal x (n) is a mean square error for the White noise (n) excited linear shift invariant system H (x) (or AR model) are available, the analysis of the signal power spectrum is estimated at:
Where: Input sequence variance, n1, a2 … …, ap for the parameters to be estimated [3-4]. functionImgZoom (Id) / / re-set the picture size to prevent bursting form (var w = $ (Id). width; var m = 650; if (w
2 Frequency Measurement System 2.1 virtual instrument front panel design
Start LabVIEW, select the option to open a new panel, and then use C0ntrols template control object (controls) and display objects (indications) to create a graphical user interface (ie front panel) [5]. In the frequency measurement system design, based on the frequency of measurement, the front panel includes signal generating module, power spectrum estimation module, the frequency measurement module and results display module of four parts, design of the interface shown in Figure 1.
We are high quality suppliers, our products such as inflatable cervical collar , medical cushion for oversee buyer. To know more, please visits finger splint@cervical collar suppliersc.
Processing your request, Please wait....
