Echo Canceller Adaptive Filter Convergence and Path Mutation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing echo cancellers face challenges in achieving effective echo suppression due to phenomena like double-talk and path mutation, which affect the convergence performance of self-adaptive filters, particularly in noisy environments and during double-end communication.
Innovation Solution
An echo canceller comprising a self-adaptive filter, a voice signal detection portion, and a path change detection portion, where a random sequence is used to initialize the filter, and voice detection thresholds are dynamically adjusted based on communication status and path changes to control the filter's operation, ensuring effective echo cancellation even during double-end communication and path mutations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional Geigel method is used for double-end detection, then the complexity is low and it is easy to be realized, but the determination of a threshold is very difficult, and the effect under the noise environment is relatively bad
Solution Approach 1:
The patent changes the detection parameters by using correlation detection instead of Geigel method, and dynamically adjusts detection thresholds based on signal characteristics to improve detection accuracy in noisy environments while maintaining manageable complexity
Solution Approach 2:
The patent implements feedback mechanisms where the detection module continuously monitors signal characteristics and adjusts detection parameters accordingly, allowing the system to adapt to changing acoustic conditions and improve detection reliability
2Measurement precision
If correlation detection method is used for double-end detection, then the detection accuracy is improved, but the complexity increases and the performance worsens when the noise is larger or the path is mutated
Solution Approach 1:
The patent makes the detection system dynamic by continuously adapting detection parameters and thresholds based on real-time signal analysis, allowing the system to maintain high accuracy while managing complexity through intelligent resource allocation
Solution Approach 2:
The patent segments the detection process into multiple stages with different detection strategies, using simpler methods when conditions permit and more complex methods only when necessary, thereby balancing accuracy and complexity
3Reliability
If self-adaptive filter continuously updates coefficients, then the echo cancellation performance is maintained, but during double-talk mode the uprush of error signals causes the self-adaptive filter to be divergent
Solution Approach 1:
The patent uses feedback control where the system monitors error signal characteristics and adjusts filter coefficient updates accordingly, reducing or pausing updates during double-talk conditions to prevent divergence while maintaining performance during normal operation
Solution Approach 2:
The patent dynamically adjusts the coefficient update mechanism based on communication status detection, switching between continuous updates and paused updates to maintain both stability and performance under varying conditions
4Speed
If the self-adaptive filter is restarted during path mutation, then the filter can rapidly track the path changes, but the initialization process increases the loss of time
Solution Approach 1:
The patent performs preliminary actions by maintaining filter state information and using warm-start techniques during path mutations, allowing the filter to resume tracking quickly without full re-initialization, thereby reducing time loss while maintaining tracking speed
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
An echo canceller and an echo cancellation method are provided. The echo canceller includes: a self-adaptive filter, a voice signal detection portion and a path change detection portion; a far-end voice signal is propagated in an echo path through a speaker and is picked up by a microphone to form an echo signal. The self-adaptive filter is configured to receive the far-end voice signal as a training signal to simulate the echo path, and cancel the echo signal in a near-end signal; the voice signal detection portion is configured to: detect a communication status, control the self-adaptive filter according to the communication status, and control startup of the path change detection portion according to the communication status; and the path change detection portion is configured to: detect whether a change occurs on the echo path, and control the self-adaptive filter according to whether the change occurs on the echo path. By adopting the above-mentioned technical solution, at an initialization stage, very good convergence effect and double-end communication effect can be achieved.