Differential Zero-Crossing Sound Recognition Power Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital sound recognition systems consume high power and pose privacy threats due to the need for continuous digital sampling and reconstruction of raw sound data, which is inefficient and insecure.
Innovation Solution
Implementing an analog-to-information (A2I) process that extracts sparse sound features directly from analog signals using ultra-low power analog or mixed signal circuitry, reducing the need for continuous digital sampling and allowing for continuous monitoring without reconstructing the original sound.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If digital sampling and reconstruction is used for sound recognition, then recognition accuracy is improved, but power consumption increases and privacy is compromised
Solution Approach 1:
The patent extracts only the essential zero-crossing count features from the analog sound signal rather than digitizing the entire signal. This extraction approach maintains recognition accuracy by capturing the most relevant acoustic information while avoiding the power consumption and privacy issues of full digital sampling and reconstruction
Solution Approach 2:
The patent applies partial action by performing only the minimum necessary digitization - converting just the zero-crossing count values to digital form rather than the entire analog signal. This partial digitization is sufficient for recognition purposes while dramatically reducing power consumption and eliminating privacy concerns associated with full signal reconstruction
2Reliability
If continuous digital sampling is performed, then sound recognition capability is maintained, but power consumption increases significantly
Solution Approach 1:
The patent extracts only the zero-crossing count feature from the continuous analog signal, converting this single scalar value to digital form rather than continuously digitizing the entire signal. This maintains sound recognition capability while enabling ultra-low power operation during idle periods
Solution Approach 2:
The patent employs periodic action by using event-triggered sampling where the analog-to-information conversion occurs only when sound events are detected, rather than continuous digital sampling. The system monitors analog zero-crossing counts periodically and only activates full processing when relevant acoustic events occur, significantly reducing average power consumption
3Loss of information
If full analog signal is digitized, then complete sound information is captured, but power consumption and data security issues arise
Solution Approach 1:
The patent extracts only the zero-crossing count feature from the analog signal, converting this compressed representation to digital form while leaving the original analog signal intact. This provides sufficient sound information for recognition purposes without creating digital copies that could compromise privacy or security
Solution Approach 2:
The patent creates a simplified digital copy of only the essential acoustic features (zero-crossing counts) rather than copying the full analog signal. This digital representation is sufficient for recognition tasks but cannot be used to reconstruct the original sound, thereby protecting privacy while maintaining information completeness for the intended application
Data Source
AI summary
A low power sound recognition sensor is configured to receive an analog signal that may contain a signature sound. Sparse sound parameter information is extracted from the analog signal and compared to a sound parameter reference stored locally with the sound recognition sensor to detect when the signature sound is received in the analog signal. A portion of the sparse sound parameter information is differential zero crossing (ZC) counts. Differential ZC rate may be determined by measuring a number of times the analog signal crosses a threshold value during each of a sequence of time frames to form a sequence of ZC counts and taking a difference between selected pairs of ZC counts to form a sequence of differential ZC counts.