Analog-to-Information Sound Sensor for Low Power Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital sound recognition systems consume high power due to continuous operation and require high-performance analog-front-end (AFE) and analog-to-digital converter (ADC) components, which also pose privacy risks by reconstructing raw input signals.
Innovation Solution
Implementing an analog-to-information (A2I) system that extracts sparse sound features directly from analog signals using ultra-low power analog or mixed signal circuitry, reducing the need for continuous digitization and minimizing power consumption while ensuring privacy through non-reconstructible feature extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continuous monitoring is performed using high-performance AFE and ADC components, then sound recognition accuracy is improved, but power consumption increases significantly
Solution Approach 1:
The patent segments the sound recognition system into two distinct parts: an analog front end (AFE) that continuously monitors audio signals with high precision, and a digital signal processor (DSP) that is activated only when the AFE detects potential events of interest. This segmentation allows the high-power DSP to remain dormant during normal operation while the low-power AFE maintains continuous monitoring capability.
Solution Approach 2:
The system implements periodic activation of the DSP based on events detected by the AFE. Instead of continuous operation, the DSP is awakened periodically or on-demand when the AFE identifies significant acoustic events, performs the necessary digital processing, and then returns to sleep mode. This periodic action dramatically reduces overall power consumption while maintaining recognition accuracy.
2Use of energy by moving object
If the sensor assembly operates in low power sleep mode, then power consumption is reduced, but the risk of missing events increases
Solution Approach 1:
The patent introduces an intermediary component - the analog front end (AFE) - that acts as a mediator between the continuous audio environment and the periodic DSP operation. The AFE continuously monitors the audio input and serves as an event detector that triggers DSP activation only when necessary, thus bridging the gap between continuous monitoring and periodic processing.
Solution Approach 2:
The AFE operates autonomously in continuous mode, self-managing the task of detecting events of interest without requiring DSP intervention. It independently evaluates incoming audio signals and autonomously triggers DSP activation only when events warrant further processing, enabling the system to maintain reliability while the DSP remains in low-power state.
3Measurement precision
If high-performance ADC components are used for continuous digitization, then sound recognition precision is improved, but power consumption and privacy risks increase
Solution Approach 1:
The patent extracts only the essential features needed for sound recognition from the continuous audio signal through the AFE, rather than digitizing and storing the entire high-fidelity audio stream. The AFE processes the analog signal to extract relevant acoustic event characteristics, which are then passed to the DSP for recognition, eliminating the need for continuous high-resolution ADC operation.
Solution Approach 2:
The system replaces expensive, high-performance continuous ADC operation with a simpler, lower-power alternative. The AFE uses basic analog processing to detect events, and only when events occur does the system activate the more resource-intensive DSP for brief periods to perform recognition on the extracted features, rather than maintaining continuous high-performance digitization.
Data Source
AI summary
A low power sound recognition sensor is configured to receive an analog signal that may contain a signature sound. The received analog signal is evaluated using a detection portion of the analog section to determine when background noise on the analog signal is exceeded. A feature extraction portion of the analog section is triggered to extract sparse sound parameter information from the analog signal when the background noise is exceeded. An initial truncated portion of the sound parameter information is compared to a truncated sound parameter database stored locally with the sound recognition sensor to detect when there is a likelihood that the expected sound is being received in the analog signal. A trigger signal is generated to trigger classification logic when the likelihood that the expected sound is being received exceeds a threshold value.