Overlapping Filter Banks for Quasi-Periodic Signal Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital representation methods for waveforms are inefficient in capturing and tracking the underlying waveform of quasi-periodic signals, leading to data redundancy and increased processing and storage requirements, especially when dealing with signals that have varying frequencies over a wide range.
Innovation Solution
A method and apparatus using a bank of overlapping filters with closely spaced center frequencies to track the strongest frequency component of quasi-periodic waveforms, performing wavelet transforms and quarter-phase decompositions to reconstruct and analyze the signal, allowing for accurate tracking of frequency changes over a large range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If high-fidelity digital representation methods are used to capture waveform signals, then signal integrity is improved, but data storage requirements and processing bandwidth increase significantly
Solution Approach 1:
The patent applies segmentation by dividing the signal processing into multiple filter banks, each handling specific frequency ranges. Instead of processing the entire signal at full resolution, the signal is segmented into frequency bands using filter banks (e.g., 32 filter banks covering different frequency ranges), and only relevant segments are processed and stored at high fidelity. This reduces overall data storage requirements while maintaining signal integrity for the frequency components of interest.
Solution Approach 2:
The patent extracts and tracks only the dominant frequency components from the signal using filter banks and spectral analysis. Rather than storing and processing all signal data at full resolution, the system extracts the essential frequency information (fundamental and harmonic components) and represents the signal in terms of these extracted features. This significantly reduces data storage requirements while preserving the essential characteristics of the waveform.
2Productivity
If traditional filter banks are used for signal decomposition, then frequency analysis is achieved, but tracking accuracy for varying frequency components deteriorates
Solution Approach 1:
The patent applies dynamics by implementing adaptive tracking mechanisms that dynamically adjust filter bank parameters based on the detected signal characteristics. The system continuously monitors the signal and adjusts the center frequencies and bandwidths of the filter banks to follow the varying frequency components. This dynamic adaptation ensures that the filter banks remain optimally tuned to the dominant frequency components even as they vary over time, thereby maintaining high tracking accuracy.
Solution Approach 2:
The patent employs feedback mechanisms where the output of spectral analysis is fed back to adjust the filter bank parameters. The system analyzes the spectral content, identifies the dominant frequency components, and uses this information to adjust the filter bank configuration for subsequent processing. This feedback loop ensures that the filter banks continuously adapt to track the varying frequency components accurately, resolving the contradiction between frequency analysis capability and tracking precision.
3Measurement precision
If closely spaced filter banks are used to track frequency components, then tracking resolution is improved, but system complexity increases
Solution Approach 1:
The patent merges multiple filter banks into a unified processing framework with shared computational resources. Instead of implementing completely separate filter banks for each frequency range, the system uses a hierarchical structure where filter banks are organized in stages (e.g., first stage with broader bands, second stage with narrower bands) and share common processing elements. This merging approach reduces the overall system complexity while maintaining the tracking resolution benefits of closely spaced filters through selective refinement in different frequency ranges.
Solution Approach 2:
The patent transitions from a one-dimensional approach (single filter bank processing) to a multi-dimensional hierarchical structure. The filter banks are organized in multiple stages with different resolution levels, creating a multi-dimensional processing architecture. This allows the system to achieve high tracking resolution where needed (closely spaced filters in critical frequency ranges) while using coarser filtering in less critical ranges, thereby reducing overall system complexity through dimensional organization of the processing resources.
Data Source
AI summary
A system and method for representing quasi-periodic waveforms, for example, representing a plurality of limited decompositions of the quasi-periodic waveform. Each decomposition includes a first and second amplitude value and at least one time value. In some embodiments, each of the decompositions is phase adjusted such that the arithmetic sum of the plurality of limited decompositions reconstructs the quasi-periodic waveform. Data-structure attributes are created and used to reconstruct the quasi-periodic waveform. Features of the quasi-periodic wave are tracked using pattern-recognition techniques. The fundamental rate of the signal (e.g., heartbeat) can vary widely, for example by a factor of 2-3 or more from the lowest to highest frequency. To get quarter-phase representations of a component (e.g., lowest frequency “rate” component) that varies over time (by a factor of two to three) many overlapping filters use bandpass and overlap parameters that allow tracking the component's frequency version on changing quarter-phase basis.