Two frameworks for tailoring signal decomposition methods

The codes will be uploaded after the article's publication.

The codes will be uploaded after the article's publication.

A naturally arising question is how to design signal decomposition methods tailored for analyzing specific kinds of signal components of interest. However, there still lacks such frameworks for tailored SD method design. To answer this question, based on the heuristic optimization that does not need gradient information and has great adaptability to various statistical optimization objective functions, our work proposed two new heuristic optimization-based signal decomposition frameworks, namely the concurrent decomposition framework and the successive decomposition framework. Once a proper statistical objective function that can quantify certain features of signal components is designed, it can be used in the proposed new frameworks to form new signal decomposition methods to extract signal components having the concerned features.