TABLE OF CONTENTS In addition to the limits on the reconstructions imposed by the Born and the Rytov approximations, there are also the following experimental limitations to consider:
Limitations caused by ignoring evanescent waves
Sampling the data along the receiver line
Finite receiver length
Limited views of the object.
Each of the first three factors can be modeled as a simple constant low pass filtering of the scattered field. Because the reconstruction process is linear the net effect can be modeled by a single low pass filter with a cutoff at the lowest of the three cutoff frequencies. The experiment can be optimized by adjusting the parameters so that each low pass filter cuts off at the same frequency.
The effect of a limited number of views also can be modeled as a low pass filter. In this case, though, the cutoff frequency varies with the radial direction.
Since evanescent waves have a complex wavenumber they are severely attenuated over a distance of only a few wavelengths. This limits the highest received wavenumber to
This is a fundamental limit of the propagation process and can only be improved by moving the experiment to a higher frequency (or shorter wavelength).
After the wave has been scattered by the object and propagated to the receiver line, it must be measured. This is usually done with a point receiver. Unfortunately, it is not possible to sample at...
