Acoustic Feedback Noise Suppression via Frequency Distortion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Adaptive feedback cancellation methods in acoustic systems often lead to signal artifacts and degradation in sound quality due to correlated input signals, particularly when dealing with noise caused by acoustic feedback, which can result in noticeable whistling noises.
Innovation Solution
The method involves frequency distorting the output signal before decorrelating it, allowing the adaptive filter to focus on target signal components while excluding noise components caused by feedback, using techniques like linear prediction and frequency shifts to adjust the decorrelation process, ensuring the adaptive filter can effectively suppress feedback noise without compromising sound quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If an adaptive filter is used to suppress feedback noise, then feedback noise suppression is improved, but sound quality deteriorates due to cancellation of target signal components
Solution Approach 1:
The patent applies frequency shifting to the output signal before it enters the adaptive filter. This preliminary frequency transformation modifies the spectral characteristics of the feedback path, allowing the adaptive filter to distinguish feedback noise from target signals more effectively. The frequency-shifted output signal is processed through the adaptive filter, which generates a compensation signal that cancels feedback noise without affecting the original target signal components.
Solution Approach 2:
The patent changes the frequency parameter of the output signal by applying a frequency shift. This parameter modification creates a frequency-offset version of the output signal that is used in the feedback path. By changing the frequency parameter, the adaptive filter can operate on a modified signal where feedback noise and target signals are more separable, thus improving noise suppression while preserving sound quality.
2Productivity
If the adaptation step size is increased to improve convergence speed, then productivity is improved, but signal artifacts increase due to divergence with correlated signals
Solution Approach 1:
The frequency shifting operation is performed in advance on the output signal before it is fed to the adaptive filter. This preliminary transformation modifies the correlation structure of the signal, reducing the harmful effects of high correlation between input signals. By preprocessing the signal with frequency shifting, the adaptive filter can converge faster with larger step sizes without experiencing divergence or generating significant signal artifacts.
Data Source
Figure 1~2
AI summary
The invention describes a method (1) for suppressing a noise (g) in an acoustic system (2), wherein the acoustic system (2) comprises at least one microphone (4) and at least one loudspeaker (6), wherein the at least one microphone (4) generates an input signal (m) and wherein the at least one loudspeaker (6) generates an acoustic signal (p) which is partially fed back to the at least one microphone (4), wherein a first intermediate signal (x) is generated along a main signal path (8) depending on the input signal (m), and an output signal (xs) is generated from the first intermediate signal (x) by a frequency distortion (22), wherein the output signal (xs) is coupled out from the main signal path (8) into a signal feedback path (16), wherein a second intermediate signal (xw) is generated from the output signal (xs) by a decorrelation (18) in the signal feedback path (16), which is used as an input for an adaptive filter (24).which generates a compensation signal (c), and wherein the compensation signal (c) is fed to the input signal (m) for compensation, wherein a third intermediate signal (ew) is formed from the input signal (m) and/or from the compensated input signal (e), which is used as an input for the adaptive filter, and wherein the output signal (xw) is fed to the at least one loudspeaker (4) for playback.