Sound Source Detection Using Autocorrelation for S/N Robustness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional sound source detection systems face challenges in robustly detecting approaching vehicles due to decreasing signal-to-noise ratios as the vehicle moves further away, leading to delayed detection and increased susceptibility to ambient noise.
Innovation Solution
The system employs autocorrelation calculations between time-series sounds collected by multiple microphones to enhance detection robustness, allowing for earlier detection of approaching vehicles by differentiating between noise and specific sound sources, and includes cross-correlation for determining the vehicle's position.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cross-correlation is used to detect sound sources from multiple microphones, then the detection direction and arrival time can be identified, but the robustness against signal-to-noise ratio decreases when the sound source is far away
Solution Approach 1:
The patent segments the sound detection process into two independent stages: first calculating autocorrelation for each microphone to detect sound presence with high S/N robustness, then calculating cross-correlation between microphones to determine direction and arrival time. This segmentation allows each stage to optimize for its specific function without being constrained by the limitations of the other approach.
Solution Approach 2:
The patent performs preliminary autocorrelation calculation on each microphone's signal before performing cross-correlation between microphones. By first identifying which microphones detect sound with high confidence using autocorrelation (which is robust to S/N ratio), the system prepares the data in advance for the cross-correlation step, ensuring that direction and arrival time calculations are based on reliable sound detection.
2Length of stationary object
If the sound source is far from the host vehicle, then the detection range is extended, but the travel sound volume decreases making detection more difficult
Solution Approach 1:
The patent replaces the conventional cross-correlation-based detection mechanism with an autocorrelation-based mechanism for the initial sound presence detection. Autocorrelation is mathematically more robust against low signal-to-noise ratios, allowing the system to detect faint sounds from distant vehicles that would be obscured by ambient noise in conventional systems.
Solution Approach 2:
The patent changes the detection parameter from cross-correlation (which compares signals between different microphones) to autocorrelation (which compares a signal with itself at different time delays). This parameter change fundamentally alters the detection characteristics, making it insensitive to ambient noise levels and enabling detection of distant sound sources with decreasing volume.
3Reliability
If autocorrelation is used instead of cross-correlation, then the robustness against signal-to-noise ratio increases, but the ability to determine sound source direction and arrival time is reduced
Solution Approach 1:
The patent segments the information extraction process into two parts: autocorrelation provides sound presence detection and timing information with high S/N robustness, while cross-correlation between microphones provides direction and precise arrival time information. By segmenting these functions, the system avoids the trade-off and achieves both robustness and position information.
Solution Approach 2:
The patent uses autocorrelation results as an intermediary step before performing cross-correlation. The autocorrelation identifies which microphones detect sound and when, providing a reliable basis for the subsequent cross-correlation that determines direction and precise arrival time. This intermediary step ensures that position information is extracted only from reliably detected sounds.
Data Source
AI summary
In a sound source detection apparatus that detects a sound source of a detection subject on the basis of collected sounds, sounds are collected by at least one sound collector, an autocorrelation between sounds collected in time series by the sound collector is calculated, and a determination as to whether or not the sound source of the detection subject exists is made on the basis of the autocorrelation. More particularly, sounds are preferably collected respectively by two or more sound collectors such that the existence of the sound source of the detection subject is determined by determining whether or not the autocorrelations of the sounds collected by the two or more sound collectors satisfy a predetermined condition. By using the autocorrelation to detect the sound source of the detection subject in this manner, high robustness against an S/N ratio is exhibited, leading to an improved detection performance when detecting the sound source of the detection subject.