Adaptive Feedforward Noise Cancellation Filter Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise-cancellation systems in vehicle cabins overreact to impulsive events like road debris strikes, leading to excessive noise cancellation and residual noise issues due to misinterpretation of impulsive events as changes in road noise.
Innovation Solution
A noise-cancellation system that uses a level detector to assess the absolute value of acceleration derivatives and sets noise signals to zero when exceeding a threshold, preventing filter coefficient adjustments during saturation or impulsive events, thereby reducing or eliminating undesirable acoustic artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the noise-cancellation system continuously adjusts filter coefficients based on accelerometer input, then noise cancellation effectiveness is improved, but acoustic artifacts are generated during impulsive events
Solution Approach 1:
The system performs preliminary detection of impulsive events by monitoring the absolute value of acceleration derivatives before they cause acoustic artifacts. When an impulsive event is detected (|d²x/dt²| > threshold), the system preemptively prevents filter coefficient adjustments, thereby avoiding the generation of acoustic artifacts while maintaining continuous noise cancellation during normal operation
Solution Approach 2:
The patent introduces an intermediary detection mechanism (level detector monitoring acceleration derivatives) between the accelerometer input and the adaptive filter coefficient adjustment. This intermediary layer identifies impulsive events and selectively blocks coefficient updates during these events, allowing the system to maintain noise cancellation effectiveness while preventing acoustic artifact generation
2Productivity
If the system responds to all acceleration changes, then noise cancellation coverage is improved, but overreaction to impulsive events occurs
Solution Approach 1:
The system applies different processing qualities to different types of acceleration events. For normal road noise, continuous filter coefficient adjustment is applied to maintain noise cancellation coverage. For impulsive events (detected when |d²x/dt²| > threshold), the system applies a different quality by preventing coefficient adjustments, thereby avoiding overreaction while maintaining comprehensive noise cancellation coverage for legitimate noise sources
3Adaptability or versatility
If filter coefficients are continuously updated, then adaptation to road noise is improved, but filter instability occurs during saturation events
Solution Approach 1:
The system performs preliminary detection of saturation events by monitoring acceleration derivative thresholds before filter instability can occur. When saturation is detected (|d²x/dt²| > threshold during saturation conditions), the system preemptively prevents filter coefficient updates, thereby maintaining filter stability while continuing to adapt to road noise during normal operating conditions
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
Effectively reduces or eliminates unwanted noise artifacts by preventing overreaction to impulsive events, maintaining effective noise cancellation while minimizing residual noise and maintaining filter stability.
Implementation Method 1
a reference accelerometer mounted to a vehicle body and arranged and configured to detect acceleration of the vehicle body
Data Source
Figure 1
Figure 2
AI summary
A system and method for reducing or eliminating undesirable acoustic artifacts when a vehicle is struck by road debris. The method includes generating a noise signal representative of a first acceleration detected by an accelerometer of a vehicle caused by a disturbance and generating a noise-cancellation signal via a controller within the vehicle. Residual noise resulting from the combination of the acoustic energy of the noise-cancellation signal and the disturbance is detected by a reference sensor, which generates a reference sensor signal based on the residual noise. The reference sensor signal is transmitted to an adaptive processing module to adapt filter coefficients. A second acceleration is detected by the accelerometer and a level detector calculates the absolute value of a derivative of the second acceleration. If the absolute value exceeds a threshold value, adjustment of the filter coefficients is prevented.