Adaptive Noise Cancellation Divergence Detection via Frequency Weighting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Adaptive noise-cancellation systems face challenges in detecting divergence, which can lead to increased noise levels instead of noise reduction, as existing divergence detection methods are inadequate in monitoring the convergence condition of adaptive filters.

Innovation Solution

A system and method for detecting divergence in noise-cancellation systems that normalize the power of error signal components, use a smoothed time gradient, and weight frequencies to determine if the system is diverging by comparing time-averaged values to thresholds, allowing for corrective actions to mitigate divergence.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing divergence detection methods are used, then the system structure remains simple, but the divergence detection precision is insufficient leading to inadequate monitoring of convergence condition

Engineering Contradiction:
Improvedivergence detection precisionVSAvoiddetection system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The error signal is decomposed into multiple frequency components, and the power spectrum is divided into different frequency bands. Each frequency component is analyzed separately to detect divergence, improving detection precision by examining specific frequency regions where divergence may occur.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The detection method transitions from analyzing the overall error signal power to examining the power spectral density across multiple frequencies. This dimensional expansion from time domain to frequency domain enables more precise detection of divergence patterns that may be obscured in the aggregate signal.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If the system continuously monitors convergence condition with high precision, then the divergence detection accuracy improves, but the computational energy consumption increases

Engineering Contradiction:
Improveconvergence condition monitoring accuracyVSAvoidcomputational energy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

Instead of uniformly monitoring all frequency components with equal precision, the method focuses computational resources on specific frequency bands where divergence is most likely to occur. The power spectral density analysis concentrates on critical frequency regions, reducing overall computational load while maintaining detection accuracy.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system performs partial monitoring by selecting specific frequency components for detailed analysis rather than processing the entire spectrum continuously. This selective approach reduces computational energy requirements while maintaining sufficient detection precision for critical divergence scenarios.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If the system responds quickly to detected divergence, then the noise amplification is prevented more effectively, but the false divergence detection may trigger unnecessary corrective actions

Engineering Contradiction:
Improvenoise cancellation reliabilityVSAvoidfalse divergence detection effects
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system continuously monitors the power spectral density of the error signal and compares it against reference levels. When divergence is detected in specific frequency bands, corrective actions are triggered. The continuous feedback loop allows the system to distinguish between transient fluctuations and genuine divergence patterns, reducing false detections while maintaining rapid response to actual divergence events.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3948845B1Systems and methods for detecting divergence in an adaptive system
Publication Date: 2023.06.07 BOSE CORP
  • EP3948845B1 patent drawingFigure 1
  • EP3948845B1 patent drawingFigure 2
  • EP3948845B1 patent drawingFigure 3

AI summary

A system for detecting divergence in a noise-cancellation system, comprising: a controller configured to: determine a power of a component of the error signal, the component being correlated to the at least one reference sensor signal; determine an average value, over a first time period, of a value representative of a time gradient of the power of the component of the error signal; and determine whether the average value is greater than a threshold.