MATLAB Exercises with COMSOL Data - Introduction to Imaging from Scattered Fields - page 219

receivers. Discuss your observations for the various scenarios and
give some speculation for reasons for the variations observed. Which
method performs best for image shape? Which method performs best
for proper scale? Please speculate as to why you think this is the case.
2. Use Cepstrum in MATLAB to generate Born and Cepstrum recon-
structions for a two squares (Choice 3) using 12 sources and 120
receivers. Use any valid permittivity. Inspect the images for the pres-
ence of aliasing. Do this as many times as necessary, decreasing the
number of receivers each time, until aliasing is present in the images.
What is the number of receivers when aliasing is first observed? Use
the method (and receiver distance) discussed in the topic to calcu-
late the maximum spacing that should be to avoid aliasing. How does
this compare with what you observed experimentally? Now re-run
the same scenario where the aliasing was first observed, except this
time set the jitter bandwidth to 2 and inspect new image for aliasing.
Do this one more time except set the jitter bandwidth to 4 and again
inspect the resulting image. Does this eliminate the aliasing from the
reconstructed image? What is the impact on the Ewald circles?
3. Use Cepstrum in MATLAB to generate Born and Cepstrum recon-
structions for any target consisting of two objects (Choice 1, 3, or 5)
using any valid permittivity and a minimum of 6 sources and 120
receivers. Be sure and show the individual Born and Cepstrum recon-
structions for each source. Please discuss your observations for the
individual Born and Cepstrum reconstructions for each source and
speculate on their relationship to the Born and Cepstrum of the com-
bined sources. Which method performs the best? Why do you think
this is?
4. Use Cepstrum in MATLAB to generate Born and Cepstrum recon-
structions for a combination of one circle, one square, and one triangle
(Choice 6) using a minimum of 12 sources and 120 receivers. Set the
permittivity equal to 1.2. Run this scenario three separate times with
everything identical with the exception of varying the filter sigma
value for sigma equal to 1, 5 (default), and 50. Observe and compare
the effects of varying sigma on each reconstruction method. Does it
affect each method the same? Why or why not? What seems to be the
optimal value for sigma for each case?
5. Use Cepstrum in MATLAB to generate Born and Cepstrum recon-
structions for a combination of one circle, one square, and one triangle
(Choice 6) using a minimum of 12 sources and 120 receivers. Set the
permittivity equal to 1.2. Run this scenario three separate times with
everything identical with the exception of varying the filter multi-
plier value to 0.01, 0.0555556 (default), and 1. Observe and compare
the effects of varying the multiplier on each reconstruction method.
Does it affect each method the same? Why or why not? What seems to
be the optimal value for this multiplier for each case?
6. Use Cepstrum in MATLAB to generate Born and Cepstrum recon-
structions for a combination of one circle, one square, and one tri-
angle (Choice 6) using a minimum of 12 sources and 120 receivers. Set
the permittivity equal to 1.2. Run this scenario three separate times
with everything identical with the exception of varying the Reference