Scribd
Upload
11 views11 pages

SystemView Signal Processing Example

This document presents an example in SystemView to demonstrate principles of sampling and aliasing, including: - Sampling a chirp signal from 0-20Hz and analyzing its bandwidth and spectrum - Implementing natural sampling by multiplying the signal with a pulse train - Designing a low-pass filter with a passband of 50Hz and stopband of 80Hz to suppress aliased spectra above the Nyquist frequency The example varies parameters like the sample rate, aperture size, and filter design to illustrate their effects on aliasing and spectrum analysis in SystemView.

Uploaded by

Bibek Boxi
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
11 views11 pages

SystemView Signal Processing Example

This document presents an example in SystemView to demonstrate principles of sampling and aliasing, including: - Sampling a chirp signal from 0-20Hz and analyzing its bandwidth and spectrum - Implementing natural sampling by multiplying the signal with a pulse train - Designing a low-pass filter with a passband of 50Hz and stopband of 80Hz to suppress aliased spectra above the Nyquist frequency The example varies parameters like the sample rate, aperture size, and filter design to illustrate their effects on aliasing and spectrum analysis in SystemView.

Uploaded by

Bibek Boxi
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd

EE521

SystemView Example
James K Beard, Ph.D.
jkbeard@[Link]
(215) 204-7932
05/09/20 SystemView Example 1
The Example
 After Sklar, Digital Communications, Second
Edition, problem 2.4 p. 101
 Principles demonstrated
 Real-world signals and their bandwidths
 Sampling and aliasing
 The effect of the sample gate width on the spectrum
of the sampled signal
 Heterodyning or mixing of real signals
 Parameters of low pass filters

05/09/20 SystemView Example 2


Why This Example is
Important
 This is a core set of issues with use of
SystemView
 Steps in solving the problem
 Select a real-world signal and determine its
bandwidth, and select a sample rate
 Implement natural sampling as multiplication by a
pulse train or other means
 Select and design a low-pass filter that provides
sufficient bandwidth yet has a stopband that
suppresses unwanted aliased signal spectra

05/09/20 SystemView Example 3


The Nyquist Bandwidth
 Bandwidths are to 40 dB down points
 About 1% is the lower limit of visual significance in
plots
 This is a conservative goal for the purposes of a
visual example
 Bottom line
 Nyquist bandwidth depends on aliasing and other
requirements
 Can be distance between 3 dB points, 10 dB points,
40 dB points, etc.
05/09/20 SystemView Example 4
The Signal
 Frequency sweep from 0 to 20 Hz
 One time only
 Implemented as Aperiodic, Custom
 Equation for chirp, logical for gate
 p(0)=sin(0.5*tpi*20*ct^2) * (ct<=1)
 Multiplication by logic provides aperiodic
gating

05/09/20 SystemView Example 5


Getting the Spectra
 Select Sink Calculator on waveform plot
 Select Complex FFT, 20*log|FFT|(dB)
 Zoom in on spectrum
 Bandwidth to 40 dB down is about 100 Hz
 Adjust simulation data rate as necessary
 Sample rate will be 100 Hz
 Default of 1 kHz is too low
 10 kHz works fine

05/09/20 SystemView Example 6


Sampling
 System sample rate of 10 kHz
 Sample rate of 100 Hz
 Sampling interval of 10 ms
 Select natural sampling aperture T of 1 ms
 System clock must be at least 2/T (why?)

 Implement as multiplication of signal by pulse


train
 Check spectrum of result

05/09/20 SystemView Example 7


LPF Filter Requirements
 Check the spectrum of the sampled signal
 Passband should be half the sample rate
 Stopband should be 40 dB down at fs-W/2
 Our example:
 Passband at 50 Hz
 40 dB down at 80 Hz
 Low Pass Filter
 Butterworth (maximally flat) selected
 Five poles needed to get stopband low enough

05/09/20 SystemView Example 8


05/09/20 SystemView Example 9
Things to vary
 The signal
 Pick your own
 Periodic or one-time
 Band-limit with LPF if necessary

 The sample rate


 The aperture size in the natural sampling
 The selection and design of the LPF
 The frequency of the sine wave

05/09/20 SystemView Example 10


Conclusions
 Sampling generates multiple replicas, or images, of the
signal separated in frequency by the sample rate
 Rolloff of the amplitude of the images increases with
aperture size
 The SystemView sample rate must be large enough to
represent the smallest features in the simulation
 Requirements for the LPF include
 Low distortion in the passband (why?)
 Passband to include signal bandwidth
 Stopband to include unwanted images
 Requirements drive selection and design of the LPF

05/09/20 SystemView Example 11

You might also like