Bilateral Microphone Noise Reduction via Minimum Power Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise reduction techniques in hearing devices are hindered by inaccurate reference signal estimates, leading to inappropriate filtering and output distortion, particularly at low Signal-to-Noise Ratios (SNR), where noise reduction is most needed.
Innovation Solution
A signal processing method that determines the minimum power in bilateral microphone signals to adjust weights for noise reduction, averaging power across channels and applying a scaling factor to minimize audible distortions, eliminating the need for complex reference signal estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If complex reference signal estimation techniques are used to improve noise reduction, then noise reduction capability is improved, but measurement precision deteriorates due to errors in Voice Activity Detection
Solution Approach 1:
The patent extracts and eliminates the problematic reference signal estimation step by using a reference-free approach. Instead of estimating noise references through Voice Activity Detection, the system directly processes microphone signals through power calculation and weighting, removing the source of measurement errors entirely
Solution Approach 2:
The patent introduces power calculation as an intermediary between raw microphone signals and final output. By calculating power values and using them to determine weights, the system creates a reliable intermediate representation that avoids the inaccuracies of direct reference signal estimation
2Reliability
If adaptive filtering is applied to eliminate noise, then noise reduction is improved, but output quality deteriorates due to target distortion
Solution Approach 1:
The patent applies partial action by using power-based weighting rather than full adaptive filtering. The weights are determined by power ratios and scaled to provide noise reduction without the excessive filtering that causes target distortion, achieving a balanced approach
Solution Approach 2:
The patent changes the operating parameters from adaptive filter coefficients to power-based weights. By calculating weights from power values and applying scaling factors, the system achieves noise reduction through parameter transformation rather than complex adaptive filtering
3Loss of information
If noise reduction processing is increased to improve SNR, then noise reduction capability is improved, but distortion in output signal increases
Solution Approach 1:
The patent implements feedback through iterative weight calculation and scaling. The system calculates weights, applies scaling factors based on power ratios, and adjusts weights to minimize distortion while maintaining noise reduction, creating a self-regulating process
Solution Approach 2:
The patent introduces dynamic weight adjustment based on instantaneous power values. The weights are continuously adapted according to the power ratios of different microphone signals, allowing the system to dynamically balance noise reduction and signal fidelity
Data Source
Figure 1
Figure 2
AI summary
A system and method for reducing unwanted sounds in signals received from an arrangement of microphones including: sensing sound sources distributed around a specified target direction by way of an arrangement of microphones to produce left and right microphone output signals; determining the power of the left and right microphone signals; determining the minimum of the two microphone power measures; and, attenuating the signals based on a comparison of the left and right microphone power measures with the minimum power measure.