Design and Optimization of Kernel Function in Time-Frequency Analysis

Zi-wen ZHOU, Guo-bao RU, Liang-cai GAN
2019 DEStech Transactions on Computer Science and Engineering  
The time-frequency distribution reflects the frequency energy distribution of the signal in different time periods, and the time-frequency analysis can effectively extract the relevant characteristic parameters of the signal. Aiming at the difficulty of regulation of time-frequency distribution kernel function, poor frequency aggregation and weak ability to suppress cross-terms, this paper proposes a kernel function with four parameters. By selecting the kernel function parameters, on the one
more » ... eters, on the one hand, the core function passband and transition band can be effectively controlled to achieve the ideal filtering effect, and on the other hand, the interference between signal components is greatly suppressed, and the time-frequency aggregation is further improved. At the same time, the kernel function parameters are optimized by using the third-order Renyi information, so that the time-frequency distribution performance and the filtering effect are optimized, which lays a foundation for effectively extracting the relevant characteristic parameters of the signal.
doi:10.12783/dtcse/cscme2019/32543 fatcat:d3jkfr2vlfh5pgiftcwkkh3tnm