Title :
Adaptive cone-kernel time-frequency analysis
Author :
Czerwinski, Richard N. ; Jones, Douglas L.
Author_Institution :
Coordinated Sci. Lab., Illinois Univ., Urbana, IL, USA
fDate :
7/1/1995 12:00:00 AM
Abstract :
Presents a technique to adaptively optimize the performance of the cone-kernel distribution (CKD) by varying the cone length in response to changing signal properties. The result is an adaptive CKD that preserves the outer hull of signal time support, yields excellent results on real signals, outperforms fixed-length CKDs, and requires only slightly more computation than a fixed-kernel distribution
Keywords :
adaptive signal processing; computational complexity; time-frequency analysis; adaptive cone-kernel time-frequency analysis; computation; cone length; cone-kernel distribution; signal properties; signal time support; Cost function; Distributed computing; Kernel; Shape control; Signal analysis; Signal resolution; Testing; Time frequency analysis; Transient analysis;
Journal_Title :
Signal Processing, IEEE Transactions on