Adaptive Filtering for Spatial Sound Reproduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Spatial sound field reproduction techniques, such as wave field synthesis and Ambisonics, face limitations in achieving high-quality spatial reproduction due to the need for a large number of loudspeakers and the imperfections inherent in Higher-Order Ambisonics, which result in blurred sound localization and reduced listening area, especially in environments with significant wall reflections.

Innovation Solution

An adaptive filtering system with multiple input and output paths, including loudspeakers and microphones, employs an iterative least mean square algorithm to adjust filter transfer functions based on error signals and psychoacoustic constraints like magnitude, pre-ringing, and post-ringing to optimize sound reproduction, effectively modeling the acoustic paths between loudspeakers and microphones.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If wave field synthesis or Ambisonics techniques are used to achieve high-quality spatial reproduction, then spatial sound field reproduction quality is improved, but the number of loudspeakers required increases significantly

Engineering Contradiction:
Improvespatial reproduction qualityVSAvoidnumber of loudspeakers
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent applies adaptive filtering techniques that dynamically adjust filter coefficients based on measured room impulse responses. By changing the parameters of the filter transfer functions through adaptive algorithms (such as LMS or RLS), the system achieves high-quality spatial reproduction with a reduced number of loudspeakers, resolving the contradiction between reproduction quality and speaker quantity

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system incorporates feedback mechanisms where microphones capture the actual sound field, error signals are generated by comparing measured responses with target responses, and these error signals drive adaptive filter adjustments. This closed-loop feedback enables the system to achieve accurate spatial reproduction with fewer speakers by continuously optimizing the filter parameters based on actual acoustic conditions

Inventive Principle:
Principle #23Feedback

2Measurement precision

If Higher-Order Ambisonics is used to improve sound field reconstruction, then spatial accuracy is improved, but localization focus becomes blurred and listening area reduces

Engineering Contradiction:
Improvesound field reconstruction accuracyVSAvoidlocalization focus quality
Core Design Contradiction:
Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent employs dynamic adaptive filtering where filter coefficients are not fixed but continuously adjusted based on real-time error signals and room acoustic conditions. This dynamic adaptation allows the system to optimize both reconstruction accuracy and localization focus by adjusting the transfer functions to match the specific acoustic environment, preventing the blurring effect that occurs with static HOA approaches

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs preliminary measurements of room impulse responses using test signals before actual sound reproduction. These preliminary measurements allow the adaptive filters to be pre-configured with appropriate transfer functions that account for room acoustics, ensuring both accurate reconstruction and sharp localization from the start without the degradation seen in conventional HOA systems

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If loudspeaker signals are determined assuming free-field conditions, then theoretical spatial accuracy is improved, but performance deteriorates in environments with wall reflections

Engineering Contradiction:
Improvetheoretical spatial accuracyVSAvoidperformance in reflective environments
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent converts the harmful effect of wall reflections into a beneficial factor by measuring the actual room impulse responses that include these reflections. The adaptive filters are then designed to compensate for and work with these reflected paths, transforming the previously harmful reflective environment into a condition that can be accurately modeled and reproduced. The system uses the reflected signals themselves as part of the transfer function modeling, achieving reliable performance in reflective environments

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The system performs preliminary measurements of the actual room acoustic conditions including all wall reflections and reverberation paths before signal reproduction. By capturing the complete impulse response with all reflective paths included, the adaptive filters are pre-configured to accurately reproduce the sound field as it actually exists in the room, rather than relying on theoretical free-field assumptions that fail in reflective environments

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2930956B1Adaptive filtering
Publication Date: 2020.07.22 HARMAN BECKER AUTOMOTIVE SYST GMBH
  • EP2930956B1 patent drawingFigure 1~2
  • EP2930956B1 patent drawingFigure 3~4
  • EP2930956B1 patent drawingFigure 5~6

AI summary

A system and method include filtering with controllable transfer functions in signal paths upstream of K ≥ 1 output paths and downstream of Q ≥ 1 source input paths, and controlling with filter control signals of the controllable transfer functions according to an adaptive control algorithm based on error signals on M ≥ 1 error input paths and source input signals on the Q source input paths. The system and method further include at least one psychoacoustic constraint.