Automatic Mother Wavelet Selection for Signal Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional signal analysis methods struggle to analyze both the time and frequency domains of a signal simultaneously, often losing information from one domain during analysis, and rely on manual selection of mother wavelets, which is inefficient.
Innovation Solution
A method and system that automatically identify an optimal mother wavelet by chopping signals into units, converting them into waveforms, scaling, and using deep architectures to compare with known wavelet families for efficient analysis, eliminating the need for feature engineering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional methods are used to analyze time domain or frequency domain separately, then analysis simplicity is maintained, but information loss occurs in the other domain
Solution Approach 1:
The signal is divided into multiple segments or frames, allowing simultaneous time-domain and frequency-domain analysis of different portions. This segmentation enables comprehensive information extraction without losing temporal or spectral characteristics.
Solution Approach 2:
The patent employs nested analysis structures where wavelet transforms are applied at multiple scales, with each scale containing further decomposed components. This nested approach allows simultaneous observation of time and frequency characteristics at different resolutions.
2Measurement precision
If manual selection of mother wavelets is performed, then analysis accuracy can be optimized, but analysis efficiency decreases
Solution Approach 1:
The system automatically selects the optimal mother wavelet by comparing signal characteristics with a library of candidate wavelets using automated criteria. This self-service mechanism eliminates manual intervention while maintaining high accuracy through systematic evaluation of wavelet-signal matching.
Solution Approach 2:
The patent changes the parameter being optimized from manual selection to automated parameter comparison. By systematically varying wavelet parameters and comparing them against signal features using computational algorithms, the system achieves both accuracy and efficiency.
3Productivity
If automated methods are used for mother wavelet selection, then analysis efficiency improves, but analysis accuracy may deteriorate
Solution Approach 1:
The automated wavelet selection process incorporates feedback mechanisms where the system evaluates the effectiveness of selected wavelets and adjusts subsequent selections based on performance metrics. This feedback loop ensures high accuracy is maintained while preserving computational efficiency.
Solution Approach 2:
The patent replaces manual mechanical selection processes with automated computational systems. By substituting human judgment with algorithmic comparison and evaluation, the system achieves both speed and accuracy through systematic, repeatable automated procedures.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Signal analysis is applied in various industries and medical field. In signal analysis, wavelet analysis plays an important role. The wavelet analysis needs to identify a mother wavelet associated with an input signal. However, identifying the mother wavelet associated with the input signal in an automatic way is challenging. Systems and methods of the present disclosure provides signal analysis with automatic selection of wavelets associated with the input signal. The method provided in the present disclosure receives the input signal and a set of parameters associated with the signal. Further, the input signal is analyzed converted into waveform. The waveforms are analyzed to provide image units. Further, the image units are processed by a plurality of deep architectures. The deep architectures provides a set of comparison scores and a matching wavelet family is determined by utilizing the set of comparison scores.