I am working in wavelet coherence analysis of two variables and I have plotted the wavelet coherence spectrum. But I wanted to find the phase lag and correlation coefficient for different time periods like 2, 2-4, 4-8,8-16, 116-32. Basically I wanted to decompose the wavelet into different time scales. I am using MATLAB for the analysis and I am pretty new in MATLAB coding. Please help me to find a solution.
The Wavelet Toolbox functions dwt / dwt2 and idwt / idwt2 can be used to analyze and synthesize wavelets in a variety of ways. MATLAB also has the waveletAnalyzer app which can be used to work with wavelets using a graphical user interface.
One option could be just using windowing or proper scaling filters and performing wavelet decomposition separately if your ultimate goal is to detect the frequency elements in a particular time period. For this option, you might use any windowing algorithm with or without overlap. Another option could be using different forms of wavelet transform. For instance, wavelet package decomposition might give you more flexibility in terms of time-frequency scaling. You might also wanna look at : Maximal Overlap Discrete Wavelet Package Transform (MODWPT) and Double Density Discrete Wavelet Transform (DWT) for better decomposition options.
For more details:
https://dsp.stackexchange.com/questions/10675/difference-between-a-wavelet-transform-and-a-wavelet-decomposition
https://www.mathworks.com/help/wavelet/ref/modwptdetails.html
https://www.mathworks.com/help/wavelet/ref/dddtree.html#:~:text=The%20double-density%20DWT%20is%20a%20three-channel%20perfect%20reconstruction,%28wavelet%29%20filters%20in%20the%20second%20and%20third%20columns.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators