background image

NI Spectral Measurements Toolkit User Guide

12

ni.com

Continuous Zoom FFT

Continuous zoom FFT is a technique for quickly analyzing data as it 
arrives. A decimation process reduces the sample rate in real time. After the 
process acquires all the data and decimates it in time 

T

, a relatively small 

FFT remains. The term 

continuous

 refers to beginning the process while 

data arrives. With a standard FFT, you must wait until all the data arrives 
before beginning calculations.

The continuous zoom FFT first shifts the spectral region of interest into 
the baseband. The technique then applies a lowpass antialias filter and 
decimates, or downsamples, the data by a factor of 

M

. The zoom factor of 

M

 yields a new effective sample rate, 

f

s

/

M

. The antialias filter has a cutoff 

frequency of 

f

s

/

(

× 

M)

 because the Nyquist frequency has decreased by 

a factor of 

M

After lowpass filtering, the continuous zoom FFT performs an FFT on the 
reduced sample rate data to produce the zoom spectrum. This technique is 
destructive because the original data changes as a result of the filtering and 
decimation. If you store the data and batch process it later, you lose the 
primary benefit of the technique, which is its real-time capability. Figure 3 
shows the basic steps of frequency shifting, decimation, and FFT.

Figure 3.  

Continuous Zoom FFT Diagram

Although it might seem possible to reduce the sample rate, 

f

s

, to improve 

frequency resolution, this method does not work. You cannot acquire the 
data with a lower sample rate to increase resolution because the Nyquist 
sampling theorem applies. In the original acquisition, you must sample at 
least twice as fast as the highest frequency of your desired zoom region to 
obtain the frequency information you need. You cannot reduce the sample 
rate until after frequency shifting occurs if you want to improve the 
frequency resolution of the zoom region. The continuous zoom FFT shifts a 
high-frequency signal into the baseband before adjusting the sample rate.

x(n)

f

s

f

s

/M

Fixed Size

FFT

Zoom Processor

Lowpass Filter

Decimate by M

Modulation

Summary of Contents for NI Spectral Measurements Toolkit

Page 1: ...es 7 SMT Spectrum Analyzer simulated 7 SMT ACP simulated 8 Using LabWindows CVI Spectral Measurements Examples 9 Continuous Zoom FFT 9 Frequency Response Measurement 9 Cross Power Spectrum Measurement...

Page 2: ...t computers such as Windows System This icon denotes a note which alerts you to important information bold Bold text denotes items that you must select or click in the software such as menu items and...

Page 3: ...the Spectral Measurements Toolkit to convert a raw FFT spectrum to a power spectrum or power spectral density for noise measurements Peak power and frequency determinations The Spectral Measurements...

Page 4: ...nts and Functions Addons Modulation palettes SMT functions are located on the LabWindows CVI function panel under Library Spectral Measurements Toolset and Library Analog Modulation Note Some Spectral...

Page 5: ...t Simulation folder The example is not intended to be executable but rather to supply an informative block diagram depicting general programming for applications using acquired or simulated data Figur...

Page 6: ...ak above the threshold level as a valid peak Advanced Level VIs The zoom processing capability of the SMT Basic Zoom Power Spectrum VI is limited by the size of the data previously acquired Use the VI...

Page 7: ...r names SMT Spectrum Analyzer simulated The SMT Spectrum Analyzer simulated VI demonstrates how to use SMT VIs to build a spectrum analyzer with zoom and averaging capabilities This VI is located in t...

Page 8: ...ks in the example spectrum In the parameter specify whether to find Single Max Peak or Multiple Peaks and enter the threshold level value above which a peak is valid The peak threshold level uses the...

Page 9: ...CVI samples smt folder You can use these example programs as a starting point for your applications The simulated folder contains examples that use a simulated source Use these examples to explore di...

Page 10: ...ndle2 when you calculate the averaged cross power spectrum between the stimulus and the response Using Spectral Measurements Toolkit Techniques The Spectral Measurements Toolkit contains tools for blo...

Page 11: ...tral region The zoom FFT uses algorithms to avoid the amount of calculation required using a long standard FFT to obtain high frequency resolution over an entire spectrum You can define the frequency...

Page 12: ...mple rate data to produce the zoom spectrum This technique is destructive because the original data changes as a result of the filtering and decimation If you store the data and batch process it later...

Page 13: ...om FFT technique This VI is located on the Zoom FFT palette which is a subpalette of the SMT Advanced palette This VI also passes the complex modulated and filtered time domain data corresponding to t...

Page 14: ...located on the Zoom FFT palette which is a subpalette of the SMT Advanced palette Determining When to Use Continuous or Block Zoom FFT Choosing which zoom FFT to use for a particular application depen...

Page 15: ...t critical use the block zoom FFT with the latest data available Zoom Spectrogram A spectrogram is the result of joint time frequency analysis JTFA processing The Spectral Measurements Toolkit impleme...

Page 16: ...Hz and the span is 2 MHz the SMT Config Zoom STFT VI calculates the spectrogram from 9 MHz to 11 MHz The time span specifies the time interval from the center of the first window to the center of the...

Page 17: ...a spectrogram with evenly distributed time and frequency resolution on a square display area If the display area is not square enter an aspect ratio for the display area in the aspect ratio parameter...

Page 18: ...frequency and span center frequency must fall within the effective band of the input signal The effective band is the frequency band in which the data from the input signal is valid You can use the ef...

Page 19: ...ates that if you enter appropriate values for both center frequency and span the values do not change The spectrum represents the frequency response of the input antialiasing filter on the data acquis...

Page 20: ...the equivalent filter is determined by the window applied to the time domain signal The resolution bandwidth control represents the width of an equivalent filter corresponding to a single FFT bin You...

Page 21: ...involving the resolution bandwidth The configuration VIs use resolution bandwidth spectral lines and window to determine the acquisition size which is the number of points that you must acquire for a...

Page 22: ...the values you need in two cases If the span and sampling frequency you need correspond to a zoom factor that is not an integer the VI coerces the zoom factor to an integer value and the span varies a...

Page 23: ...Table 1 shows the trade offs of using two different RBWs Note An RBW of 103 5 kHz is not sufficient to distinguish between two tones that are 100 kHz apart Figure 9 RBW Example Table 1 Larger versus...

Page 24: ...minate the effect of phase noise The magnitude of the spectrum is independent of time shifts of the input signal but the phase can change dramatically with each data record If you average the power sp...

Page 25: ...y independently to each frequency bin of the Fourier transform FR jFI The complex representation of the Fourier transform of a signal f t using real and imaginary values F k FR jFI For vector averagin...

Page 26: ...where Yk is the new average Yk 1 is the previous average and Xk is the new measurement For linear weighting a1 k 1 k and a2 1 k For exponential weighting a1 k 1 k and a2 1 N for k N a1 N 1 N and a2 1...

Page 27: ...verage the most recent N measurements Continuous Average all measurements taken with equal weight Averaged FFT Spectrum The following equations describe the three averaging methods applied to a comple...

Page 28: ...ross spectrum Sxy results from multiplying the complex conjugate of spectrum X by spectrum Y as follows Sxy conj X Y For RMS averaging an averaged cross spectrum consists of the average of the individ...

Page 29: ...pectrum in many ways depending on the nature of the measured signal and the aspect of the signal that you need to quantify The SMT Spectrum Unit Conversion VI converts and scales a spectrum to the rep...

Page 30: ...s parameter includes the following subparameters window determines the ENBW of the window you use The ENBW affects the spectral density calculations because of the spectral leakage effect of windowing...

Page 31: ...If the VI does not apply a window a finite sample size is equivalent to rectangular windowing If you specify which window the VI applies to the input signal the VI uses a curve fitting algorithm on t...

Page 32: ...annel Power ACP The SMT Adjacent Channel Power VI located on the SMT Measurements palette measures the way a center channel and its two adjacent channels distribute power Use this VI before using the...

Page 33: ...the total power of the signal For a specified percentage B the upper and lower limits of the frequency band are the frequencies above and below which 100 B 2 of the total power is found This measurem...

Page 34: ...has offices located around the world to help address your support needs For telephone support in the United States create your service request at ni com support and follow the calling instructions or...

Page 35: ...of their respective companies For patents covering National Instruments products refer to the appropriate location Help Patents in your software the patents txt file on your media or ni com patents 2...

Reviews: