Analog systems for the spectral analysis and signal processing /
In this dissertation analog systems for spectral analysis and
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1996.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=739667851&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | In this dissertation analog systems for spectral analysis and signal processing are explored. In the area of signal analysis two tools for computing spectral information are investigated, the variable center frequency bandpass filter and the wavelet correlator. The main characteristics that we look for in these tools are their flexibility to compute the frequency components of a signal for a given resolution, low hardware overhead, and fast processing. The research in this area provide the theoretical foundations and the procedure to design hardware capable of efficiently performing the spectral analysis tasks. Experimental results are obtained to substantiate the theoretical findings as well as establishing their limitations. For spectral signal processing the capabilities of cellular neural networks (CNN) are explored, specially for applications in subband coding for image compression. For the last few years CNN has been developed as a universal image processor. The development of algorithms for analog CNN hardware is an important research task. On the other hand, image compression is one of the most important operations in image processing that has not been implemented with CNN. Wavelet coding is one of the most successful techniques in compression, providing high compression ration while preserving its quality. This research attempts to fulfill these needs. To process one-dimensional signals a novel CNN structure is investigated. CNN has been extensively used in image processing, but little attention has been given to one- dimensional signals. The main objective of this research is to explore the possible applications of the CNN paradigm in this field. The main applications considered are in wavelet decomposition and noise reduction. |
|---|---|
| Item Description: | Vita. "Major Subject: Electrical Engineering". |
| Physical Description: | xi, 152 leaves : illustrations ; 28 cm. Issued also on microfiche from University Microfilms Inc. |
| Bibliography: | Includes bibliographical references: pages 129-135. |