Vehicle Noise Cause Identification with Microphone Response Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise generation cause identification methods in vehicles suffer from inaccuracies due to variations in microphone models, leading to deviations in frequency responses and reduced accuracy in identifying sound-generating portions.
Innovation Solution
A noise generation cause identifying method and device that corrects the frequency response of sound signals using model information to align with a learning microphone's response, employing multiple response correcting processes and variable obtaining methods to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a map that has undergone machine learning is used to estimate a portion acting as the cause of a sound generated in a vehicle, then the identification process is automated and efficient, but the accuracy of identifying sound-generating portions deteriorates due to variations in microphone models and frequency response deviations
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing correction values for frequency response deviations in a correction value storage unit. These correction values are computed in advance for different microphone models and frequency bands, allowing the system to quickly retrieve and apply appropriate corrections during runtime without performing complex real-time calculations, thus maintaining both efficiency and accuracy
Solution Approach 2:
The patent changes the parameter of frequency response characteristics by applying model-specific correction values to normalize the frequency responses of different microphone models. The system identifies the actual microphone model, retrieves corresponding correction values, and adjusts the sound signal parameters to compensate for manufacturing variations, thereby improving identification accuracy while maintaining automated processing
2Measurement precision
If correction values for frequency response deviations are stored and applied based on microphone models, then the accuracy of sound source identification is improved, but the device complexity and data storage requirements increase
Solution Approach 1:
The patent segments the frequency spectrum into multiple frequency bands and stores correction values separately for each band and microphone model combination. This segmentation allows the system to apply targeted corrections only where needed rather than processing the entire frequency spectrum uniformly, reducing the computational burden and organizing the correction data in a manageable, structured manner
Solution Approach 2:
The patent creates simplified copies of frequency response characteristics in the form of pre-computed correction values. Instead of storing complete frequency response curves or performing complex real-time analysis, the system uses compact correction value tables that capture the essential deviations, significantly reducing storage requirements and processing complexity while maintaining correction effectiveness
3Measurement precision
If the frequency response of sound signals is corrected to align with a learning microphone's response, then variations in frequency response are reduced and identification accuracy is improved, but the processing time and computational load increase
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing correction values for frequency response deviations in a correction value storage unit. These correction values are computed in advance for different microphone models and frequency bands, allowing the system to quickly retrieve and apply appropriate corrections during runtime without performing complex real-time calculations, thus maintaining both efficiency and accuracy
Solution Approach 2:
The patent implements dynamic correction by adapting the processing approach based on the identified microphone model. The system dynamically selects the appropriate correction values from storage based on the actual microphone model detected, applying only the necessary corrections rather than uniformly processing all signals, thereby optimizing processing time while maintaining accuracy
Data Source
AI summary
A noise generation cause identifying method and a noise generation cause identifying device are provided. A response correcting process corrects a sound signal obtained through a sound signal obtaining process based on obtained model information so that a frequency response of the obtained sound signal approaches a frequency response of a learning sound signal. A variable obtaining process obtains a variable output from a map by inputting the corrected sound signal to the map. A cause identifying process identifies a generation cause of a sound picked up by a microphone using the variable obtained through the variable obtaining process.