Resonator Array Audio Sensing for Low-Power Frequency Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio sensing technologies face challenges in efficiently acquiring frequency domain information due to the power-intensive Fourier transformation process, which is burdensome and not suitable for continuous monitoring in always-ready devices like mobile phones and smart homes, especially when distinguishing audio signals from surrounding noise.

Innovation Solution

An audio sensing device with a resonator array is developed, where a plurality of resonators are arranged on a membrane within a vacuum cavity, allowing for direct mechanical sensing of sound frequencies without the need for Fourier transformation, reducing power consumption and enabling fast, real-time monitoring of frequency domain information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If Fourier transformation is used to acquire frequency domain information, then frequency analysis capability is improved, but power consumption and computational burden increase

Engineering Contradiction:
Improvefrequency analysis capabilityVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The audio sensing device divides the frequency spectrum into multiple bands, with each resonator beam tuned to a specific frequency band. Instead of performing a complete Fourier transformation of the entire audio signal, the device segments the frequency analysis into parallel resonator beams that can be processed independently and simultaneously, reducing computational burden and power consumption.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent replaces the computational Fourier transformation process with a mechanical resonance-based frequency separation system. Resonator beams with different natural frequencies mechanically filter and respond to different frequency components of the audio signal, substituting mathematical computation with physical resonance phenomena to achieve frequency domain analysis with lower power consumption.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If Fourier transformation is performed continuously for always-ready state monitoring, then frequency domain information availability is improved, but computational burden and power consumption increase

Engineering Contradiction:
Improvefrequency domain information availabilityVSAvoidcomputational burden
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The audio sensing device divides the audio signal processing into multiple parallel channels, each corresponding to a resonator beam tuned to a specific frequency band. This segmentation allows continuous monitoring of frequency domain information across different bands simultaneously, maintaining always-ready state capability while distributing the computational burden across multiple simple parallel operations rather than one complex sequential transformation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The resonator beams naturally respond to audio signals at their resonant frequencies through periodic vibration when exposed to corresponding frequency components. This periodic mechanical response enables continuous frequency domain monitoring without requiring continuous computational Fourier transformations, reducing the complexity of always-ready state monitoring while maintaining frequency information availability.

Inventive Principle:
Principle #19Periodic action

3Measurement precision

If resonator beams are arranged in a plane above a membrane, then frequency sensing capability is improved, but device area increases

Engineering Contradiction:
Improvefrequency sensing capabilityVSAvoiddevice area
Core Design Contradiction:
Measurement precisionVSArea of stationary object

Solution Approach 1:

The resonator beams are arranged in a compact planar configuration above the membrane, with each beam nested within the overall device footprint. This nested arrangement allows multiple frequency sensing elements to be integrated in a small area, maintaining frequency sensing capability across multiple bands while minimizing the total device area through efficient spatial utilization.

Inventive Principle:
Principle #7Nested doll (Nesting)

Solution Approach 2:

The patent utilizes the vertical dimension by positioning resonator beams above the membrane in a planar array rather than spreading them out horizontally. This three-dimensional arrangement allows multiple resonator beams to be stacked or arranged in layers above the membrane surface, increasing frequency sensing capability without proportionally increasing the horizontal device area.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

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 solution allows for low-power, fast acquisition of frequency domain information in an always-ready state, effectively removing noise and enabling real-time monitoring, while integrating multiple resonators for various frequency bands in a small area, suitable for applications in voice recognition, context awareness, and noise reduction.

Implementation Method 1

a plurality of resonators having different resonant frequencies are arranged on a membrane

Methodology Applied
Scientific EffectResonance: Resonance

Implementation Method 2

the plurality of resonators are disposed inside a cavity and an interior of the cavity is maintained in a vacuum state

Methodology Applied
Scientific EffectVacuum: Vacuum

Data Source

PatentEP2986024B1Audio sensing device and method of acquiring frequency information
Publication Date: 2019.05.01 SAMSUNG ELECTRONICS CO LTD
  • EP2986024B1 patent drawingFigure 1
  • EP2986024B1 patent drawingFigure 2
  • EP2986024B1 patent drawingFigure 3

AI summary

An audio sensing device having a resonator array and a method of acquiring frequency information using the audio sensing device are provided. The audio sensing device includes a substrate having a cavity formed therein, a membrane provided on the substrate and covering the cavity, and a plurality of resonators provided on the membrane and respectively sensing sound frequencies of different frequency bands.