Adaptive Filter Step Size Adjustment for Hearing Device Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional adaptive feedback cancellation algorithms in ear-wearable devices face challenges in adapting quickly to perturbations caused by user movements and environmental changes, leading to instability and reduced performance.
Innovation Solution
The implementation of an adaptive feedback canceller with a machine learning-based instability detector that extracts features from error signals to dynamically adjust the step size of the adaptive foreground filter, allowing for faster adaptation to perturbations and improved stability detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a traditional adaptive feedback cancellation algorithm is used, then the device structure remains simple, but the adaptation speed to perturbations is slow
Solution Approach 1:
The patent implements dynamic adaptation by switching between two different adaptive filter algorithms based on detected instability conditions. The system transitions from a first adaptive filter algorithm under normal conditions to a second adaptive filter algorithm when instability is detected, enabling the system to adapt its processing behavior dynamically rather than using a fixed algorithm throughout operation.
Solution Approach 2:
The patent changes key parameters of the adaptive filter system by switching between different algorithms with distinct characteristics. The second adaptive filter algorithm uses different step sizes and update rules compared to the first algorithm, fundamentally changing the adaptation parameters to achieve faster convergence during unstable conditions.
2Speed
If the step size of the adaptive filter is increased to speed up adaptation, then the convergence rate improves, but the stability of the error signal deteriorates
Solution Approach 1:
The system dynamically adjusts the step size parameter by switching between two different adaptive filter algorithms. During stable operation, a first algorithm with smaller step sizes maintains error signal stability. When instability is detected, the system switches to a second algorithm with larger step sizes to accelerate convergence, thus dynamically optimizing the step size based on real-time conditions.
Solution Approach 2:
The system employs periodic monitoring of error signal characteristics through an instability detector that continuously analyzes the error signal. Based on this periodic assessment, the system alternates between different adaptive filter configurations, switching to faster adaptation modes when instability patterns are detected and returning to stable modes when conditions improve.
3Adaptability or versatility
If a fixed adaptive filter algorithm is used, then the device operation is simple, but the system cannot quickly respond to environmental changes and user movements
Solution Approach 1:
The system transitions from a fixed algorithm approach to a dynamic multi-algorithm system. An instability detector continuously monitors error signal characteristics and triggers switching between different adaptive filter algorithms when environmental changes or user movements cause feedback path variations, enabling the system to adapt its behavior dynamically to changing conditions.
Solution Approach 2:
The system uses feedback from an instability detector that analyzes error signal characteristics to determine when to switch between adaptive filter algorithms. This feedback mechanism allows the system to respond to environmental changes and user movements by detecting instability patterns and automatically adjusting the adaptation strategy accordingly.
Data Source
AI summary
An adaptive feedback canceller of an ear-wearable device has an adaptive foreground filter that inserts a feedback cancellation signal into a digitized input signal to produce an error signal. An instability detector of the device is configured to extract wo or more features from the error signal. The instability detector has a machine learning module that determines instability in the error signal based on the two or more features. The instability module changes the adaptive foreground filter in response to determining the instability. The change causes the adaptive foreground filter to have a faster adaptation to perturbations in the error signal compared to a previously used step size.