Audio Precompensation Filter for Bright-Zone Sound Without Pre-Ringing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio precompensation methods, such as Acoustic Contrast Control (ACC) and Energy Difference Maximization (EDM), fail to effectively manage sound distribution in shared spaces, leading to pre-ringing and other acoustical artefacts due to their focus on acoustical contrast without considering bright zone properties and causality of filters.
Innovation Solution
A method and system for determining filter parameters of an audio precompensation filter that optimizes a criterion function under constraints of causality and stability, incorporating terms for deviation in the bright zone and weighted differences in power between bright and dark zones to reproduce a desired sound field while minimizing sound in undesired areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ACC or EDM methods are used to maximize acoustical contrast, then acoustic separation between bright and dark zones is improved, but pre-ringing and acoustical artefacts occur due to excessively long signal build-up times
Solution Approach 1:
The patent applies preliminary action by pre-compensating the audio signal with a causal filter before it reaches the loudspeakers. The filter parameters are determined in advance to achieve the desired acoustic contrast and bright zone properties, preventing pre-ringing artefacts by ensuring the filter is causal (does not require future signal values). This pre-compensation approach allows the system to achieve good acoustic separation without the harmful pre-ringing effects that plague non-causal optimization methods.
2Reliability
If ACC or EDM methods are used to maximize acoustical contrast, then acoustic separation is improved, but bright zone sound quality is neglected
Solution Approach 1:
The patent applies local quality by optimizing the criterion function to independently control both the acoustic contrast between zones and the sound quality within the bright zone. The criterion includes separate terms: one for acoustic contrast (difference between bright and dark zone energies) and another for bright zone properties (deviation from target sound field). This allows different quality requirements to be met in different spatial locations, with the bright zone receiving optimized sound quality treatment while the dark zone receives suppressed sound.
3Ease of manufacture
If filter parameters are optimized without causality constraints, then mathematical tractability is improved, but the resulting filters are non-causal and produce pre-ringing artefacts
Solution Approach 1:
The patent applies preliminary action by incorporating causality constraints directly into the filter parameter optimization process. Rather than optimizing without constraints and then attempting to fix the resulting non-causal filter, the causality constraint is built into the criterion function from the beginning. This ensures that the optimization yields a causal filter that can be implemented in real-time without producing pre-ringing artefacts, while still achieving good acoustic contrast and bright zone properties.
4Manufacturing precision
If existing precompensation methods are used, then sound field reproduction is achieved, but they do not consider spatial distribution of sound in shared spaces
Solution Approach 1:
The patent applies another dimension by extending the optimization criterion from traditional single-zone sound field reproduction to multi-zone spatial control. The criterion function incorporates terms for both bright zone sound field reproduction (matching target sound field) and dark zone sound suppression (minimizing energy). This adds a spatial distribution dimension to the optimization, allowing the system to control sound in multiple simultaneous zones rather than treating the entire space uniformly.
Data Source
AI summary
Disclosed is a system configured to determine filter parameters of an audio precompensation filter for the compensation of an associated sound generating system having at least two loudspeakers. The system is configured to obtain sound field models describing the sound field in at least one region of space designated the bright zone and the sound field in at least one region of space designated the dark zone. The system is also configured to obtain a target sound field in the bright zone. Further, the system is configured to determine filter parameters of the audio precompensation filter so that a criterion function is optimized under the constraint of causality and stability of the dynamics of the audio precompensation filter to enable reproduction of a desired target sound field in the bright zone, while reproducing as little sound as possible in the dark zone.