Multi-Sensor Acoustic Signal Processing for Speaker Feedback Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Acoustic systems face issues of acoustic feedback, such as howling and echo, due to the formation of closed-loop circuits where sound emitted by a speaker is recaptured by a sound sensor and amplified, causing discomfort and interference.
Innovation Solution
An acoustic system with M sound sensors and a signal processing circuit that performs filtering and synthesis operations on sound pickup signals using M sets of target filtering parameters to minimize the speaker's sound component in the composite signal, reducing or eliminating feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the sound sensor collects and amplifies ambient sound including speaker output, then the hearing compensation function is achieved, but acoustic feedback and howling occur due to the closed-loop circuit
Solution Approach 1:
The patent extracts and removes the harmful speaker sound component from the ambient sound signal captured by the sound sensor. The signal processing circuit identifies and eliminates the feedback path by subtracting the speaker output signal from the captured ambient sound, thereby removing the closed-loop feedback that causes howling while preserving the useful hearing compensation function.
Solution Approach 2:
The patent converts the harmful feedback signal into a useful cancellation signal. By capturing the speaker output through the sound sensor and processing it to create an anti-phase signal, the system transforms the problematic feedback loop into a beneficial active noise cancellation mechanism that eliminates howling while maintaining amplification of desired ambient sounds.
2Adaptability or versatility
If the speaker plays remote user voice through the local speaker, then communication functionality is achieved, but echo interference occurs when the sound is recaptured by the local sound sensor
Solution Approach 1:
The patent extracts the remote user's voice signal that was played through the local speaker from the combined signal captured by the local sound sensor. By identifying and separating this specific component, the system removes the echo path while preserving the local user's voice and other ambient sounds for accurate transmission to the remote end.
Solution Approach 2:
The patent introduces an intermediary signal processing operation that mediates between the speaker output and the sound sensor input. The processing circuit acts as an intermediary by generating a cancellation signal based on the known speaker output and subtracting it from the sensor capture, thereby eliminating echo interference without affecting the communication functionality.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution effectively reduces or eliminates feedback sound, preventing howling and echo in acoustic systems by minimizing the speaker's sound component in the composite signal while maintaining the accuracy of the target sound source signal.
Implementation Method 1
a speaker, configured to receives a driving signal and converts the driving signal into a first sound
Implementation Method 2
a sound sensor module, where the sound sensor module comprises M sound sensors and is configured to pick up an ambient sound and generate M sound pickup signals
Data Source
AI summary
A signal processing method and an acoustic system are provided. When M sound sensors in a sound sensor module are in operation, they collect an ambient sound and generate M sound pickup signals. The ambient sound comprises a first sound from a speaker and a second sound from a target sound source. A signal processing circuit can perform a filtering operation on the M sound pickup signals based on M sets of target filtering parameters to obtain M filtered signals, and perform a synthesis operation on the M filtered signals to obtain a composite signal, and then perform a target operation on the composite signal. Since the M sets of target filtering parameters are configured to minimize a signal component from the speaker in the composite signal under a target constraint.