Digital Signal Processing Reference
In-Depth Information
Figure 20.13
Segment (zoomed) of the reconstructed signal sample value sequence overlap-
ping the waveform of the original signal
Note that while the approach to data compression-signal reconstruction de-
scribed above is applicable for a wide variety of signals, the achievable recon-
struction quality to a certain extent depends on spectra of them. To achieve good
signal reconstruction adaptability to specific sampling irregularities, it should
be possible to decompose the respective signals into their components and the
frequencies of these components should be estimated with sufficiently high res-
olution and precision. That can usually be done but there might be problems.
20.5
Reducing the Quantity of Sensors in
Large-aperture Arrays
The subject of high-performance complexity-reduced array signal processing is
rather complicated and is actually beyond the scope of this topic. Therefore only
a few points on this subject are considered here, simply to draw attention to
the fact that the discussed digital alias-free signal processing technology is quite
competitive and has a high application potential in this area.
As explained in Chapter 17, pseudo-randomization of the distances between
sensors in arrays helps in the suppression of the aliasing effect and under certain
conditions might lead to a significant reduction in the number of sensors in arrays.
However, to succeed, it is crucial to use appropriate nonuniform signal process-
ing techniques for handling array signal processing, both in the time and spatial
