| Summary: | 碩士 === 國立海洋大學 === 電機工程學系 === 85 === In diverse fields of application, such as telecommunicatios,
telemetry, sonar, and radar,the received signals are
generally nonstationary, and they must be
processed by appropriate time-frequency analysis methods
for obtaining genuine information.
For multicomponent nonstationary signals, each time-frequency
distribution corresponds to a kernel that controls the
cross-terms suppression properties. Selection of a fixed
kernel limits the class of signals for which the distribution
performs well.For analysis of a broad class signals, a signal-
dependent kernel is recommended.
In this thesis, a new procedure, based on optimal criteria and
Gaussian kernel, is introduced for signal-dependent kernel
design.We use Matlab and LabVIEW to build up the
processing for conducting simulation and
experiments with various nonstationary chirpsignals. The
simulation and experimental results show the
algorithm is efficient.
|
|---|
