Harmonic Echo Power Estimator for Loudspeaker Nonlinearities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing echo cancellation systems in communication equipment struggle to effectively remove nonlinear echoes from harmonic loudspeaker nonlinearities due to their reliance on linear models, which lead to reduced transparency and increased computational complexity when trying to estimate and suppress these nonlinearities.
Innovation Solution
A method and system for estimating and suppressing echo power from harmonic loudspeaker nonlinearities using a frequency selective filter based on the ratio between near-end and echo signal power estimates, involving a harmonic echo power estimator that maps loudspeaker output signal frequency bands to input signal bands, determines power estimates, and transforms these estimates across the echo path to accurately calculate nonlinear echo power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a linear model is used for echo subtraction, then computational simplicity is achieved, but nonlinear echoes from harmonic loudspeaker nonlinearities cannot be removed
Solution Approach 1:
The invention segments the echo cancellation task into two parts: linear echo removal using the existing linear model, and nonlinear echo removal using a separate nonlinear model. The nonlinear model specifically targets harmonic distortions by modeling the loudspeaker's nonlinear behavior separately from the acoustic echo path, allowing both linear and nonlinear echoes to be removed effectively without increasing overall system complexity
Solution Approach 2:
The invention introduces a nonlinear model as an intermediary component that specifically handles harmonic nonlinearities. This nonlinear model acts as a mediator between the linear echo subtractor and the residual echo suppressor, providing accurate nonlinear echo estimates that enable the residual suppressor to work more effectively with reduced suppression levels
2Measurement precision
If a Volterra model is used to model harmonic loudspeaker nonlinearities, then nonlinear echo estimation is achieved, but computational complexity increases significantly
Solution Approach 1:
The invention extracts only the essential nonlinear characteristics needed for echo cancellation - specifically the harmonic distortion components - rather than using a complete Volterra model. By taking out only the relevant nonlinear terms and modeling them in a simplified manner, the system achieves adequate nonlinear echo estimation without the excessive computational burden of a full Volterra implementation
Solution Approach 2:
The invention changes the modeling parameters from a comprehensive Volterra model with many coefficients to a simplified nonlinear model with fewer parameters. By parameterizing the nonlinear model in terms of loudspeaker input signal characteristics and harmonic relationships rather than full impulse response coefficients, the computational complexity is reduced while maintaining estimation accuracy
3Reliability
If the residual echo suppressor performs extra suppression to compensate for uncertainty in nonlinear echo power estimate, then echo attenuation requirements are met, but transparency of near-end signal is reduced
Solution Approach 1:
The invention implements feedback by using the nonlinear model to continuously estimate nonlinear echo power and feeding this information back to the residual echo suppressor. This feedback mechanism allows the suppressor to adapt its suppression level dynamically, applying only the necessary amount of suppression to meet echo attenuation requirements while preserving near-end signal transparency through accurate real-time nonlinear echo power estimation
Data Source
Figure 1
Figure 2A~3
Figure 4
AI summary
Described is a harmonic echo power estimator configured to estimate power of echo generated by harmonic loudspeaker nonlinearities in a user equipment having an echo path between a loudspeaker input and a microphone output. This estimator includes a frequency band mapper (40) configured to map each frequency band in a set of loudspeaker output signal frequency bands (b isp ) into a corresponding array of loudspeaker input signal frequency bands (b(b isp ,k)), where each frequency band in the set is mapped into several frequency bands in the corresponding array. A power estimator (42) is configured to determine a power estimate (P x (b(b isp ,k))) of each input signal in each array of frequency bands. A power estimate combiner (44) is configured to combine determined power estimates (P x (b(b isp ,k))) in each array of frequency bands into a corresponding estimate (P x, nl (b isp )) of loudspeaker input power generating harmonic loudspeaker nonlinearities. A power estimate transformer (46) is configured to transform the estimates of loudspeaker input power across the echo path (EP) into power estimates (P s, nl (b)) of the echo generated by the harmonic loudspeaker nonlinearities.