Hearing Aid Channel Allocation for Feedback-Prone Frequency Bands
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital hearing aid devices face challenges in efficiently managing frequency bands to minimize computational power consumption and reduce acoustic feedback, while maintaining effective sound processing and resolution.
Innovation Solution
A hearing aid device with a frequency band bundling and allocation scheme that dynamically or statically bundles input frequency bands based on the likelihood of feedback, using a matrix to determine which bands to process in fewer channels, and redistributes them to maintain high frequency resolution in feedback-prone regions, thereby reducing computational power and energy consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of input frequency bands is increased to maintain high frequency resolution, then the frequency resolution is improved, but the computational power consumption increases
Solution Approach 1:
The patent divides the frequency spectrum into multiple input frequency bands (e.g., 64 bands) and processes them through multiple processing channels. This segmentation allows high frequency resolution to be maintained in the input stage while enabling subsequent bundling operations to reduce the number of active processing channels, thereby balancing frequency resolution with computational power consumption.
Solution Approach 2:
The patent implements frequency band bundling where multiple adjacent input frequency bands are bundled into fewer processing channels. For example, multiple input frequency bands are grouped and processed together in a reduced number of processing channels (e.g., 16 channels), reducing computational load while maintaining the ability to redistribute to the original number of output frequency bands for high resolution output.
2Use of energy by moving object
If frequency bands are bundled to reduce computational effort, then energy consumption is decreased, but the frequency resolution may be degraded
Solution Approach 1:
The patent employs dynamic frequency band bundling where the bundling configuration is not fixed but can be adapted based on listening conditions, feedback characteristics, and user needs. The system can dynamically adjust which frequency bands are bundled together and to what extent, allowing optimal balance between energy consumption and frequency resolution to be maintained under varying operational conditions.
Solution Approach 2:
The patent applies different bundling strategies to different frequency regions based on their specific characteristics. Frequency bands prone to feedback or requiring higher resolution can be processed with less bundling or excluded from bundling, while other regions can undergo more aggressive bundling. This localized approach ensures that frequency resolution is maintained where critical while achieving energy savings where possible.
3Power
If gain is increased to improve hearing amplification, then the hearing assistance is improved, but acoustic feedback is more likely to occur
Solution Approach 1:
The patent utilizes feedback detection to identify frequency bands where acoustic feedback is occurring or likely to occur. Instead of simply reducing gain in these regions, the system uses the feedback information to intelligently configure frequency band bundling, directing bundled processing away from feedback-prone regions while maintaining high gain in safe regions. This converts the harmful feedback phenomenon into useful information for optimizing the bundling configuration.
Solution Approach 2:
The patent implements a feedback detection mechanism that continuously monitors for acoustic feedback and uses this information to adaptively adjust the frequency band bundling configuration. When feedback is detected in certain frequency bands, the system modifies the bundling to process those bands differently, thereby maintaining high amplification gain overall while suppressing feedback in specific regions through intelligent channel allocation.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The invention relates to a hearing aid comprising a first microphone configured to receive a first acoustic signal and to convert the first acoustic signal to a first electrical audio signal, a speaker configured to emit an acoustic output signal into an ear of a user of the hearing aid device, a first analog-to-digital converter for converting the first electrical audio signal into a first time domain input signal, a first input unit comprising a first analysis filter bank which is configured to convert the first time domain input signal to a number NI,1 of first input frequency bands wherein the number NI,1 of first input frequency bands is determined by said first analysis filter bank, a first frequency band bundling and allocation unit which is configured to bundle adjacent first input frequency bands and to allocate first frequency bands to be processed to a number NP,1 of first processing channels, a memory unit which is configured to store data indicating which of the first NI,1 input frequency bands are subject to a likelihood of feedback that is above a threshold, a signal processing unit is configured to process the first frequency bands to be processed in the number NP,1 of first processing channels, and wherein the number NP,1 of first processing channels is smaller than the number NI,1 of first input frequency bands, and wherein the first frequency band bundling and allocation unit is configured to generate a first bundling and allocation scheme which determines the bundling of the first NI,1 input frequency bands and the allocation of the first frequency bands to be processed to the first NP,1 processing channels wherein said first bundling and allocation scheme depends on the likelihood of feedback to occur in at least one of the first NI,1 input frequency bands.