Análisis espectral de wavelet: una concisa revisión

• Autores: Josué Moisés Polanco Martínez
• Localización: Revista de Climatología, ISSN-e 1578-8768, Nº. 22, 2022, págs. 51-95
• Idioma: español
• Enlaces
  • Texto completo (pdf)
• Resumen
  • español

    En este artículo se presenta una concisa revisión de una de las herramientas más utilizadas en el procesamiento de señales, el análisis espectral de wavelet mediante la transformada continua de wavelet para los casos uni- y bi-variados. Esta herramienta es sumamente útil para analizar cualquier tipo de datos que no sean estacionarios, esto es, aquellos datos cuyas principales caractarísticas estadísticas (como la media o la varianza) cambian con el tiempo.

  • English

    This article presents a short but concise review of one of the most used techniques in signal processing, the wavelet spectral analysis (WSA) via the continuous wavelet transform (CWT) for the uni- and bi-variate cases. This mathematical tool is very useful to analyse any kind of data that are not stationary, that is, data that their main statistical properties (such as the mean or the variance) can change with time. The range of applications of this methodology is enormous, ranging from experimental sciences (such as physics or engineering) to social sciences (such as economics or archaeology). However, and despite the fact that the examples presented in this review article are related to environmental sciences, all the theoretical concepts as well as the computational aspects to use this technique are applicable to any area of science or engineering. Additionally, this review article presents some freely distributed computer programs (of the free software or open source type) that perform spectral analysis via the continuous wavelet transform.