Microphone Array Noise Localization and Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In enclosed areas such as vehicles and buildings, identifying and diagnosing abnormal sounds is challenging due to numerous variables affecting sound reproduction, making it difficult to pinpoint and correct noise issues reported by passengers.
Innovation Solution
Deploying on-board microphone array devices that capture and store sound data, using direction of arrival information and artificial intelligence systems for analysis, including trained model recognition and neural networks, to identify and localize noise sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional manual noise reporting methods are used, then passengers can report abnormal sounds, but the maintenance crew cannot efficiently identify and diagnose the noise sources due to numerous variables affecting sound reproduction
Solution Approach 1:
The system continuously records audio data in a circular buffer before noise events occur, so that when a passenger reports an abnormal sound, the data is already captured and stored for immediate analysis. This eliminates the need to try to reproduce the noise under various operating conditions, as the actual noise data is preserved for later examination by maintenance personnel.
Solution Approach 2:
The system creates digital copies of the actual noise events as they occur, storing them in memory for later analysis. These audio copies preserve the exact characteristics of the noise including frequency, amplitude, and temporal patterns, allowing maintenance personnel to analyze the noise without needing to reproduce the operating conditions that generated it.
2Measurement precision
If multiple variables such as yaw, pitch, roll, g-forces, turbulence, temperature, and engine rpm are considered to reproduce noise conditions, then more complete diagnostic information may be obtained, but the complexity and time required for noise identification increases significantly
Solution Approach 1:
The system continuously monitors and records not only audio data but also associated operational parameters (such as engine rpm, temperature, and other sensor data) in the circular buffer before noise events occur. This preliminary capture of contextual data eliminates the need to later reproduce complex operating conditions, as all relevant variables are already recorded alongside the noise for comprehensive analysis.
3Quantity of substance
If the circular buffer continuously erases and rewrites data, then memory space is efficiently utilized, but the ability to store complete noise event information may be compromised
Solution Approach 1:
The system continuously pre-records audio data in the circular buffer before noise events occur, ensuring that when a passenger reports an abnormal sound, the buffer already contains the noise event plus surrounding context. This preliminary action ensures that even though the buffer is continuously overwriting old data, the specific noise event of interest is preserved because recording continues continuously and the report triggers a save of the relevant window.
Solution Approach 2:
The system uses passenger reports as feedback to determine which portions of the continuously recorded data should be preserved and transferred to long-term storage. When a passenger reports a noise event, the system identifies the corresponding time window in the circular buffer and saves that specific segment, ensuring that the reported noise is captured even as other data is overwritten.
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 pinpoints and categorizes abnormal sounds, providing accurate diagnosis and potential causes, improving the efficiency of noise issue resolution by analyzing data before, during, and after the event, and offering insights into Mean Time Between Failure (MTBF) of components.
Implementation Method 1
The microphone array devices each employ multiple microphone transducers, in a predefined physical arrangement, so they can be used to capture direction of arrival information used to pinpoint where every captured sound is coming from
Data Source
AI summary
A sound pickup transducer array, deployed within an enclosed area, is coupled to a sound recorder. A processor, coupled to the sound recorder, provides a button or speech recognizer through which a person in the enclosed area issues a command signifying the occurrence of a sound for which categorizing is requested. The processor is programmed to respond to the issued command by extracting and storing an audio snippet copied from the audio recorder, in a digital memory, where the snippet corresponds to sound captured before, during and after the issued command. The processor communicates the stored audio snippet to an artificial intelligence system trained to categorize sounds as to what produced them. The artificial intelligence system may employ trained model feature extraction, a neural network categorization system, and/or direction of sound arrival analysis.