Signal Feature Extraction Using Extended Impulse Response Filters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In filter bank analysis, signal features leak out to adjacent band-pass filters due to overlapping bands, leading to redundant analysis results and difficulty in analyzing output signals effectively.
Innovation Solution
The technique generates an extended impulse response matrix and calculates expansion coefficients using an input signal vector and the impulse response sequence of band-pass filters, outputting signals that correspond to center impulse response vectors to reduce leakage between adjacent filters and account for diverse feature patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If filter bank analysis using overlapping band-pass filters is applied, then signal feature extraction is achieved, but signal leakage occurs to adjacent filters causing redundant analysis results
Solution Approach 1:
The patent extracts only the necessary signal components by calculating expansion coefficients that represent the input signal as a linear combination of impulse response sequences. This selective extraction approach isolates the dominant filter responses while suppressing leakage to adjacent filters, thereby achieving precise feature extraction without redundant information from overlapping bands.
Solution Approach 2:
The patent changes the parameter representation by transforming the filter bank output from direct time-domain signals to expansion coefficients in a transformed domain. This parameter transformation allows for better separation of signal components and reduces the harmful leakage effect while preserving the essential feature information for analysis.
2Adaptability or versatility
If multiple band-pass filters with overlapping bands are used, then comprehensive signal analysis is achieved, but analysis results become redundant and difficult to interpret
Solution Approach 1:
The patent introduces a feedback mechanism where the calculated expansion coefficients are used to reconstruct the signal and compare it with the original input signal. This feedback loop ensures that only the necessary filter responses are retained while redundant information from overlapping bands is suppressed, simplifying the interpretation of analysis results while maintaining comprehensive signal coverage.
Solution Approach 2:
The patent segments the signal analysis by dividing it into distinct expansion coefficient calculations for each filter, then combining them through a structured linear combination. This segmentation approach organizes the complex multi-filter analysis into manageable components, making the results easier to interpret while maintaining comprehensive signal analysis coverage.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The purpose of the present invention is to reduce leakage of an output signal between band-pass filters and in the time axis direction. A signal input part 1 vectorizes an input signal x(n). An extended impulse response matrix generation part 2 generates an extended impulse response matrix He in which impulse response vectors using impulse response sequences of band-pass filters as elements are extended in the time axis direction. An expansion coefficient calculation part 3 calculates an expansion coefficient vector ^y(n) using an input signal vectors -x(n) and the extended impulse response matrix He. A signal output part 4 outputs at least one of expansion coefficients corresponding to a center vector of the extended impulse response matrix He of the expansion coefficient vector ^y(n).