Acoustic Echo Cancellation Using Null Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional acoustic echo cancellers require knowledge of the far-end signal strength and magnitude to effectively cancel echoes, and they typically need two acoustic echo cancellers for a two-microphone system, which is inefficient.
Innovation Solution
The solution involves predicting between two microphones and placing a null in the direction of the loudspeaker, using a null coefficient adapted without requiring knowledge of the far-end signal's strength or magnitude, allowing for echo cancellation between a loudspeaker and two microphones without access to far-end signal information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional acoustic echo cancellers are used for each microphone, then echo cancellation can be achieved, but the system complexity increases and efficiency decreases
Solution Approach 1:
The patent merges the echo cancellation functionality for multiple microphones into a single acoustic echo canceller that processes signals from multiple microphones simultaneously. Instead of implementing separate AEC instances for each microphone, the system uses one unified AEC that handles all microphone inputs, thereby reducing overall system complexity while maintaining effective echo cancellation across all channels.
Solution Approach 2:
The acoustic echo canceller is designed to perform multiple functions by processing signals from multiple microphones through a single device. This universal approach allows one AEC to serve multiple purposes and handle multiple microphone inputs, eliminating the need for dedicated AEC units for each microphone and improving system efficiency.
2Measurement precision
If far-end signal strength and magnitude information is used, then echo cancellation accuracy improves, but the system requires additional information that may not be available
Solution Approach 1:
The system uses the near-end microphone signals themselves to adapt the filter coefficients for echo cancellation, rather than relying on far-end signal information that may not be available. The AEC leverages the characteristics of the near-end signals and the derived transmission path to perform adaptive filtering and echo cancellation independently, making the system self-sufficient without requiring external far-end signal data.
Solution Approach 2:
Instead of using far-end signal information to cancel echoes, the system inverts the approach by using near-end microphone signals and deriving the transmission path characteristics from them. The adaptation process works backwards from the near-end observations to achieve echo cancellation, eliminating the dependency on far-end signal availability.
3Reliability
If adaptive filtering is performed between loudspeaker and each microphone, then echo cancellation is effective, but the computational load and system resources increase
Solution Approach 1:
The patent combines the adaptive filtering operations for multiple microphones into a single unified process. Instead of performing separate adaptive filtering for each microphone-loudspeaker pair, the system executes one adaptive filtering operation that processes all microphone signals simultaneously, thereby reducing computational load and energy consumption while maintaining effective echo cancellation.
Data Source
AI summary
Systems and methods for providing acoustic echo cancellation are provided. Primary and secondary acoustic signals are received by a communication device. The acoustic signals may include loudspeaker leakage. A null coefficient is then adaptively determined for each subband of the secondary acoustic signal. The null coefficient is applied to the secondary acoustic signal to generate a coefficient-modified signal. The coefficient-modified signal is subtracted from the primary acoustic signal to generate a masked acoustic signal with reduced or no echo. The masked acoustic signal may be output.