A method to align segments of time series based on envelope features as anchor points

C. JARNE; P. N. ALCAIN

Anales AFA

Asociación Física Argentina

Lugar: Tandil; Año: 2019 vol. 30 p. 1 - 4

0327-358X

In the eld of time series analysis, there is not a unique recipe for studying signal similarities. When having the repetition of a pattern, averaging diferent signals of the same nature could be complicated. Sometimes averaging is essential in the analysis of the data. Here we propose a method to align and average segments of time series with similar patterns. For this procedure, a simple implementation based on python code is provided. This analysis was inspired by the study of canary sound syllables, but it is possible to apply itin semi-periodic signals of different nature, not necessarily related to sounds.