Proximate Dual MEMS Microphones for Low EIN Speech Reception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional MEMS microphones suffer from high equivalent input noise (EIN) levels, which degrade the signal-to-noise ratio (SNR) and make it difficult to understand speech, especially at higher frequencies, and require additional signal processing for directional sensitivity.
Innovation Solution
A proximate dual microphone system where two MEMS microphones are placed in close acoustical proximity, typically within 5 mm of each other, to generate in-phase and coherent signals that increase the speech signal strength by 6 dB while keeping EIN increase to only 3 dB, thereby improving SNR and reducing EIN effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If conventional MEMS microphones are used, then the device size and cost are reduced, but the equivalent input noise (EIN) level increases
Solution Approach 1:
The patent divides a single microphone function into two separate MEMS microphones working in parallel. Each microphone captures the sound signal independently, and their outputs are combined through signal processing to achieve lower effective EIN while maintaining the compact size advantage of MEMS technology.
Solution Approach 2:
The patent combines the outputs of two MEMS microphones through coherent signal summation. By placing the microphones in close proximity and processing their signals to maintain phase coherence, the system merges the useful sound signal while the incoherent EIN from each microphone averages out, resulting in a net reduction of effective noise.
2Ease of manufacture
If conventional MEMS microphones are used, then manufacturing simplicity is maintained, but signal-to-noise ratio (SNR) deteriorates
Solution Approach 1:
The system uses two independent MEMS microphones that can be manufactured using standard MEMS processes, maintaining ease of manufacture. The segmentation allows each microphone to be produced separately with conventional techniques while their combined output improves SNR through noise averaging.
Solution Approach 2:
The patent changes the system-level parameter of SNR by combining multiple microphone signals rather than changing the individual microphone characteristics. This allows standard MEMS microphones to be used while achieving improved SNR through the mathematical combination of multiple measurements.
3Device complexity
If single MEMS microphone is used, then device complexity is low, but speech intelligibility at higher frequencies deteriorates
Solution Approach 1:
The patent segments the audio capture function into two microphones to preserve high-frequency speech components that are often masked by EIN in single-microphone systems. The segmentation enables frequency-dependent noise reduction where high-frequency speech cues are preserved while EIN is reduced through incoherent averaging.
Solution Approach 2:
The system creates a redundant copy of the acoustic signal through a second microphone. This copying allows the system to preserve weak high-frequency speech components by averaging multiple measurements, effectively filtering out the random EIN while maintaining the useful speech information.
4Measurement precision
If directional sensitivity is achieved through beamforming, then speech signal directionality is improved, but device complexity and processing requirements increase
Solution Approach 1:
The patent extracts only the essential element for directional sensitivity—the phase and amplitude relationship between two closely-spaced microphones—while discarding the complex beamforming processing. By using microphones in very close proximity, the system achieves sufficient directional information through simple signal combination without requiring sophisticated beamforming algorithms.
Solution Approach 2:
Instead of implementing full beamforming processing, the patent uses a partial approach with two microphones in close proximity that provides sufficient directional sensitivity for many applications. This excessive simplicity (using more microphones than the minimum one) achieves the desired directional capability without the complexity of advanced signal processing.
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
The proximate dual microphone system enhances speech intelligibility, particularly in far-field applications, by maintaining an omnidirectional polar pattern and reducing EIN by 3 dB, resulting in improved SNR and sensitivity without the need for complex beamforming techniques.
Implementation Method 1
MEMS microphones are variants of the condenser microphone design. In a MEMS microphone, a pressure-sensitive diaphragm can be etched directly into a silicon wafer by MEMS processing techniques
Implementation Method 2
Microphones are transducers that convert sound energy into an electrical signal
Implementation Method 3
two MEMS microphones are placed in close acoustical proximity, typically within 5 mm of each other, to generate in-phase and coherent signals that increase the speech signal strength by 6 dB while keeping EIN increase to only 3 dB, thereby improving SNR and reducing EIN effectively
Data Source
AI summary
In one embodiment of an audio system, the audio system comprises signal summation of two MEMS microphones in close acoustical proximity for speech frequencies to provide various benefits and improvements, including improvements to: speech intelligibility, signal-to-noise ratio, effective equivalent input noise, and at-a-distance acoustic signal reception.