| Autor: |
J. Solano González, F. Garcia Nocetti, E. Rubio Acosta, E. Moreno Hernández |
| Rok vydání: |
2000 |
| Předmět: |
|
| Zdroj: |
IFAC Proceedings Volumes. 33:169-174 |
| ISSN: |
1474-6670 |
| DOI: |
10.1016/s1474-6670(17)35465-4 |
| Popis: |
Conventional methods for signal analysis utilise the Fourier Transform to estimate the spectral response of a signal. However this current practice suffers from poor frequency resolution when estimating non-stationary signals. This paper presents some alternative methods based on time-frequency distributions such as the Wigner Ville, the Choi Williams, the Bessel and the Born Jordan from a Cohen's class point of view. For each case, a continuous and discrete distribution is formulated, a criterion for determining the interaction between the spectral components of the signal is given and the simplified discretised expression for the calculation of the distribution is proposed that can produce a reduction of at least half of the computations realised when using the original time-frequency distribution definition. A general parallel architecture for the parallel computation of the distribution is also proposed. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
