Adaptive Multi-Microphone Beamforming Without Location Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional microphone beamforming techniques require accurate tracking of a user's location and direction of arrival, which is challenging in noisy environments and when microphones are not aligned in a straight line, leading to inefficiencies in hands-free communication and voice-controlled devices.
Innovation Solution
An adaptive multi-microphone beamforming method that uses a normalized least mean square (NLMS) based adaptive filter to align audio signals without requiring calculations for the user's location or direction of arrival, allowing for arbitrary microphone placement and improved signal-to-noise ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional Delay-Sum beamforming is used to enhance speech signal, then speech amplification is achieved, but accurate tracking of user location and direction of arrival is required which becomes difficult in noisy environments and when microphones are not aligned
Solution Approach 1:
The patent extracts and removes the complex location tracking and direction of arrival calculation components from the traditional beamforming system. By using adaptive filtering to directly process microphone signals without requiring explicit user location tracking, the system eliminates the problematic subsystem while maintaining speech enhancement capability
Solution Approach 2:
The patent replaces the mechanical/geometric approach of Delay-Sum beamforming (which relies on precise microphone alignment and geometric location tracking) with an adaptive signal processing approach using NLMS filtering. This substitution allows the system to achieve beamforming effects without the rigid structural requirements
2Ease of manufacture
If microphones are placed in arbitrary positions to simplify device design, then ease of manufacture is improved, but traditional beamforming requires microphones to be aligned in a straight line
Solution Approach 1:
The patent introduces dynamic adaptation through the NLMS algorithm that automatically adjusts filter coefficients based on the actual microphone positions and acoustic environment. This dynamic approach replaces the static geometric alignment requirement, allowing microphones to be placed arbitrarily while the system adapts to their actual positions
Solution Approach 2:
The patent changes the fundamental parameters of the beamforming approach by moving from fixed geometric delay calculations to adaptive filter coefficient adjustment. This parameter change enables the system to accommodate arbitrary microphone placements by continuously optimizing the filtering parameters based on actual signal characteristics
3Measurement precision
If user location tracking is implemented to achieve accurate beamforming, then speech direction accuracy is improved, but system complexity and computational requirements increase
Solution Approach 1:
The patent implements a self-service approach where the adaptive filter automatically adjusts its coefficients based on the incoming signals from microphones at arbitrary positions. The system serves itself by adapting to the actual acoustic environment and microphone configuration without requiring external location tracking or complex preprocessing
Solution Approach 2:
The patent introduces an adaptive filter as an intermediary component that processes signals from microphones at arbitrary positions. This intermediary translates the arbitrary microphone inputs into aligned speech signals without requiring direct knowledge of user location, effectively mediating between the physical microphone arrangement and the beamforming output
Data Source
AI summary
Provided is a method and computer program product for producing an enhanced audio signal for an output device from audio signals received by 2 or more microphones in close proximity to each other. For example, one embodiment of the present invention comprises the steps of receiving a first input audio signal from the first microphone, digitizing the first input audio signal to produce a first digitized audio input signal, receiving a second input audio input signal from the second microphone, digitizing the second input audio input signal to produce a second digitized audio input signal, using the first digitized audio input signal as a reference signal to an adaptive prediction filter, using the second digitized audio input signal as input to said adaptive prediction filter and finally adding a prediction result signal from the adaptive prediction filter to the first digitized audio input signal to produce the enhanced audio signal. In other embodiments, any number of microphones can be used, and in all embodiments there is no requirement to detect or locate the source or direction of arrival of the input audio signals.