Road Noise Cancellation Filter Adaptation for Transient Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Road noise cancellation systems mis-adapt due to non-stationary events such as driving over train tracks or hitting potholes, leading to degraded noise cancellation performance.
Innovation Solution
Implement a method to detect non-stationary events by analyzing sensor data using Fast Fourier Transform (FFT) and modify the adaptation parameter, such as reducing the step size of the Least Mean Square (LMS) algorithm, to prevent mis-adaptation of controllable filters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the LMS RNC system continuously adapts W-filters based on acceleration inputs and microphone signals, then the system can optimally cancel steady-state road noise, but the system mis-adapts when encountering transient non-stationary events such as driving over train tracks or hitting potholes
Solution Approach 1:
The patent implements a feedback mechanism where the error signal from the microphone is used to continuously adjust the W-filters through the LMS algorithm. The system monitors the cancellation performance and adapts the filter coefficients to minimize the error signal, creating a closed-loop control system that maintains optimal noise cancellation for steady-state conditions.
Solution Approach 2:
The patent makes the adaptation parameter (step size) dynamic by adjusting it based on the detected signal characteristics. When transient events are detected through signal analysis, the step size is reduced to prevent mis-adaptation. This dynamic adjustment allows the system to be highly adaptive during normal conditions while becoming stable during transient events, resolving the contradiction between adaptability and reliability.
2Adaptability or versatility
If the W-filters are adapted based on transient non-stationary events, then the system attempts to cancel these specific signals, but the RNC performance worsens for a period of time after the events as the system needs to re-adapt
Solution Approach 1:
The patent performs preliminary action by detecting transient events before they cause significant mis-adaptation. Through continuous signal analysis of the acceleration inputs and error signals, the system identifies non-stationary events early and proactively reduces the adaptation parameter, preventing the mis-adaptation from occurring in the first place rather than correcting it afterward.
Solution Approach 2:
The patent implements periodic signal analysis at defined frames to detect transient events. The system continuously monitors the spectral characteristics of the input signals at regular intervals, allowing it to periodically assess whether transient events are occurring and adjust the adaptation parameter accordingly, creating a rhythm of detection and response that prevents performance degradation.
3Productivity
If the LMS algorithm uses a larger step size for faster adaptation, then the system can quickly converge to optimal W-filters, but the system becomes more susceptible to mis-adaptation from transient events
Solution Approach 1:
The patent makes the step size dynamic rather than fixed. During steady-state conditions, a larger step size enables fast convergence and quick adaptation to changing road conditions. When transient events are detected through signal analysis, the step size is reduced to prevent mis-adaptation. This dynamic adjustment of the adaptation parameter allows the system to achieve both fast adaptation speed and stability against transient events at different times.
Solution Approach 2:
The patent changes the adaptation parameter (step size) based on the detected signal characteristics and operating conditions. By monitoring the spectral content and variability of the input signals, the system adjusts the step size parameter to match the current situation, using larger values for rapid convergence during normal operation and smaller values during transient events to maintain stability.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A road noise cancellation (RNC) system may include a signal analysis controller for detecting non-stationary, transient events based on sensor signals having a spectral or temporal character significantly different from steady-state road or cabin noise. Upon detection of such non-stationary events, the RNC system may modify the sensor signals to mask the non-stationary event, thereby preventing the RNC system's adaptive filters from mis-adapting because of transient, non-stationary events. Alternatively, the RNC system may pause or slow or pause adaptation of its controllable filters for the duration of a frame that includes the non-stationary event.