Noise-Cancellation Sensor Screening for Corrupted Signal Exclusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Noise-cancellation systems face performance degradation due to corrupted sensor signals, which can occur from loose wiring connections or other causes, leading to 'crackling' sounds and requiring effective detection and exclusion to maintain system efficacy.
Innovation Solution
A noise-cancellation system and method that utilize sensors to determine the power spectral density of sensor signals at various frequencies, calculate a measure of association using correlation coefficients, and exclude signals exceeding a predetermined threshold to generate a noise-cancellation signal, while notifying users of corrupted sensors and disengaging them to prevent performance degradation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor signals are used for noise-cancellation, then noise reduction performance is improved, but corrupted sensor signals cause performance degradation and crackling sounds
Solution Approach 1:
The system performs preliminary detection of corrupted sensor signals by analyzing the power spectral density and calculating correlation coefficients before the signals are used for noise-cancellation. This advance detection allows the system to identify and exclude corrupted signals proactively, preventing performance degradation and crackling sounds from occurring during operation.
Solution Approach 2:
The system introduces an intermediary detection and analysis mechanism that sits between the sensor signals and the noise-cancellation processing. This intermediary layer analyzes the power spectral density and correlation coefficients to determine signal quality, acting as a filter that allows only clean signals to proceed to the noise-cancellation algorithm.
2Productivity
If all sensor signals are processed for noise-cancellation, then noise reduction coverage is improved, but corrupted signals affect overall system performance
Solution Approach 1:
The system applies local quality control by individually assessing each sensor signal's quality through power spectral density analysis and correlation coefficient calculation. Instead of treating all signals uniformly, the system identifies which specific signals are corrupted and excludes only those from processing, while continuing to process clean signals for comprehensive noise reduction coverage.
Solution Approach 2:
The system segments the sensor signals into separate quality-assessed streams, dividing them into corrupted and non-corrupted categories. This segmentation allows the system to process only the clean signals through the noise-cancellation algorithm, maintaining high noise reduction coverage while preventing corrupted signals from degrading overall performance.
3Reliability
If sensor signal quality is monitored and corrupted signals are excluded, then noise-cancellation performance is maintained, but system complexity increases
Solution Approach 1:
The system replaces complex mechanical or hardware-based signal filtering mechanisms with computational methods. By using algorithms to calculate power spectral density and correlation coefficients, the system achieves sophisticated signal quality assessment through software processing rather than requiring complex physical filtering hardware, thereby managing complexity through information processing rather than mechanical complexity.
Data Source
AI summary
A noise-cancellation system, including: a plurality of sensors, each sensor outputting a sensor signal; a controller configured to receive each sensor signal, and, for each sensor signal, to: determine a power of the sensor signal at a plurality of frequencies; determine a measure of association between the power of the sensor signal at the plurality of frequencies and frequency; and determine whether the measure of association exceeds a predetermined threshold, wherein the processor is further configured to compute a noise-cancellation signal using the sensor signals, wherein the noise-cancellation signal is computed excluding sensor signals that were determined to exceed the predetermined threshold; and at least one actuator receiving the noise-cancellation signal and producing a noise-cancellation audio signal.