Acoustic Source Localization Using Spatial Masks and Beamforming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to effectively track and map acoustic sources in complex environments using ultra large microphone arrays, as they do not adaptively select subsets of microphones and process signals efficiently, especially in scenarios with multiple sources, reflections, and interference.
Innovation Solution
The system employs a processor to determine spatial masks with pass and rejection regions, selecting subsets of microphones and beamforming filters that maximize gain in pass regions while minimizing interference, using optimization criteria that do not depend on the acoustic source, allowing for adaptive processing of signals across the environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ultra large arrays of microphones are used to monitor acoustic environment, then source recognition and source separation are improved, but device complexity and computational requirements increase significantly
Solution Approach 1:
The patent divides the ultra large microphone array into multiple subsets, each associated with a specific spatial mask and beamforming filter. This segmentation allows the system to process signals from different spatial regions independently, reducing the computational complexity of handling all microphones simultaneously while maintaining source recognition accuracy.
Solution Approach 2:
The patent applies different beamforming filters to different subsets of microphones based on their spatial locations and the specific acoustic sources being monitored. Each subset is optimized for its local region, allowing the system to achieve high measurement precision for source recognition without requiring uniform complex processing across the entire array.
2Loss of information
If all microphones in ultra large arrays are processed simultaneously, then complete acoustic scene coverage is achieved, but computational complexity increases
Solution Approach 1:
The patent segments the microphone array into multiple subsets, each responsible for monitoring specific spatial regions defined by pass region masks. This allows the system to achieve complete acoustic scene coverage by combining results from multiple subsets while avoiding the computational burden of processing all microphones simultaneously in a single unified process.
Solution Approach 2:
The patent processes only the necessary subset of microphones for each specific acoustic source or region of interest at any given time, rather than continuously processing all microphones. This partial action approach maintains complete scene coverage through multiple scans while reducing instantaneous computational complexity.
3Productivity
If adaptive subset selection of microphones is implemented, then processing efficiency is improved, but device complexity increases
Solution Approach 1:
The patent pre-defines multiple spatial masks and associates specific beamforming filters with each mask before actual acoustic monitoring begins. This preliminary setup of subsets and filters allows the system to quickly switch between different processing configurations based on the current acoustic scene, improving processing efficiency without requiring complex real-time adaptive selection algorithms during operation.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables efficient detection and tracking of acoustic sources, characterizing them as targets or interferences, and creates an acoustic map, improving signal-to-noise ratio and reducing computational complexity, even with large numbers of microphones.
Implementation Method 1
each subset being related with a beamforming filter that maximizes the gain of the pass region mask and minimizes the gain for the complementary rejection masks
Data Source
AI summary
Systems and methods are provided to create an acoustic map of a space containing multiple acoustic sources. Source localization and separation takes place by sampling an ultra large microphone array containing over 1020 microphones. The space is divided into a plurality of masks, wherein each masks represents a pass region and a complementary rejection region. Each mask is associated with a subset of microphones and beamforming filters that maximize a gain for signals coming from the pass region of the mask and minimizes the gain for signals from the complementary region according to an optimization criterion. The optimization criterion may be a minimization of a performance function for the beamforming filters. The performance function is preferably a convex function. A processor provides a scan applying the plurality of masks to locate a target source. Processor based systems to perform the optimization are also provided.


