Dynamic Beamformer for Acoustic Echo Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High acoustic coupling between microphones and loudspeakers in close proximity leads to poor duplex quality and susceptibility to audible artifacts in audio communication channels, as existing echo cancellers assume a linear acoustic coupling path and fail to effectively cancel loudspeaker distortion.
Innovation Solution
A dynamic beamformer is used in conjunction with acoustic echo cancellation, dynamically modifying its beampattern to reduce acoustic coupling by optimizing for different communication states, such as single-user or dual-user talk scenarios, and employing multiple beampatterns to maximize signal-to-noise ratio and minimize echo cancellation penalties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If microphones and loudspeakers are placed in close proximity to reduce device size, then device compactness is improved, but acoustic coupling between microphones and loudspeakers increases causing poor duplex quality and audible artifacts
Solution Approach 1:
The patent segments the acoustic signal processing into multiple independent components: far-end signal processing path and near-end signal processing path. The far-end signal is processed separately through echo generation and cancellation, while the near-end signal is processed through beamforming and distortion compensation. This segmentation allows the system to handle the acoustic coupling problem by treating different signal paths independently, thereby maintaining device compactness while reducing the harmful effects of acoustic coupling between microphones and loudspeakers.
Solution Approach 2:
The patent introduces several intermediary processing components to mediate the acoustic coupling problem. These include the beamformer that processes near-end microphone signals, the acoustic echo canceller that removes far-end echo, and the near-end distortion compensator that specifically targets loudspeaker distortion. These intermediaries act as signal processing layers that isolate and compensate for the harmful acoustic coupling effects, allowing compact device design while maintaining audio quality.
2Device complexity
If acoustic echo cancellers assume a linear acoustic coupling path, then algorithm simplicity is improved, but cancellation performance deteriorates due to inability to cancel loudspeaker distortion
Solution Approach 1:
The patent implements dynamic processing by introducing a near-end distortion compensator that actively adapts to and compensates for loudspeaker distortion in real-time. This dynamic component works in conjunction with the beamformer and acoustic echo canceller to handle non-linear acoustic coupling effects. The system dynamically adjusts its processing based on the detected distortion characteristics, thereby improving echo cancellation performance without requiring overly complex static algorithms.
Solution Approach 2:
The near-end distortion compensator serves as an intermediary component that specifically addresses the non-linear distortion introduced by loudspeakers. This compensator is positioned in the signal processing chain to receive the beamformed near-end signal and apply distortion compensation before the final output. By introducing this specialized intermediary, the system can handle non-linear acoustic coupling effects without completely redesigning the entire echo cancellation algorithm, thus balancing complexity and performance.
3Device complexity
If a single beampattern is used for beamformer processing, then processing simplicity is improved, but audio quality deteriorates due to inability to optimize for different communication states
Solution Approach 1:
The patent implements dynamic beampattern selection by monitoring the communication state (such as near-end talker detection, far-end talker detection, and double-talk conditions) and switching between different beampatterns accordingly. During near-end talk, a beampattern optimized for near-end speech enhancement is selected, while during far-end talk, a different beampattern optimized for echo cancellation is used. This dynamic adaptation allows the system to maintain high audio quality across different communication scenarios without requiring overly complex processing for all possible states simultaneously.
Solution Approach 2:
The patent changes the beamformer parameters (specifically the beampattern coefficients) based on the detected communication state. By dynamically adjusting these parameters, the system optimizes audio quality for different scenarios: enhancing near-end speech during near-end talk, suppressing echo during far-end talk, and maintaining balanced performance during double-talk. This parameter adaptation allows the system to achieve high audio quality without the need for a single overly complex fixed beampattern.
Data Source
AI summary
Near-end equipment for a communication channel with far-end equipment. The near-end equipment includes at least one loudspeaker, at least two microphones, a beamformer, and an echo canceller. The communication channel may be in one of a number of communication states including Near-End Only state, Far-End Only state, and Double-Talk state. In one embodiment, when the echo canceller determines that the communication channel is in either the Far-End Only state or the Double-Talk state, the beamformer is configured to generate a nearfield beampattern signal that directs a null towards a loudspeaker. When the echo canceller detects the Near-End Only state, the beamformer is configured to generate a farfield beampattern signal that optimizes reception of acoustic signals from the near-end audio source. Using different beamformer processing for different communication states allows echo cancellation processing to be more successful at reducing echo in the signal transmitted to the far-end equipment.