Real-Time Ultrasonic Bat Call Detection via Selective Sample Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bat detectors using time expansion methods are not suitable for real-time monitoring of bat activity due to the need to buffer and drain ultrasonic sound samples over long periods, preventing continuous monitoring of bat activity.
Innovation Solution
A method that selects a fraction of input samples based on criteria such as root-mean-squared power levels to produce output samples at a slower rate, allowing for real-time playback of ultrasonic animal sound signals while preserving harmonic structure and amplitude details, suitable for live or recorded bat calls and other short-duration signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If time expansion detectors buffer ultrasonic sound samples at high sample rate and drain the buffer at slower sample rate, then spectral and amplitude information is preserved, but real-time monitoring capability is lost due to long buffer drainage periods
Solution Approach 1:
The patent divides the ultrasonic signal processing into segments by organizing samples into frames and selectively choosing representative samples from each frame. This segmentation allows the system to process and output samples in real-time while preserving the essential spectral characteristics of the original ultrasonic signal, resolving the contradiction between measurement precision and real-time monitoring capability.
Solution Approach 2:
The patent applies partial action by selecting only a fraction of the input samples (specifically, samples with highest root-mean-squared power levels) rather than processing all samples. This selective sampling approach maintains the essential spectral information while enabling real-time output, thus resolving the contradiction between preserving spectral detail and achieving real-time monitoring.
2Loss of information
If all ultrasonic samples are processed and output at reduced rate, then complete spectral information is maintained, but background noise and processing complexity increase
Solution Approach 1:
The patent applies local quality by differentiating between important and unimportant samples within each frame. By calculating root-mean-squared power levels and selectively choosing samples with highest energy content, the system preserves spectral information locally in the most significant samples while discarding redundant low-energy samples, thereby reducing processing complexity without losing essential information.
Solution Approach 2:
The patent extracts only the most significant samples from each frame based on their power levels, rather than processing all samples. This extraction of essential information reduces the overall processing complexity and data volume while maintaining the complete spectral information needed for accurate bat call detection and analysis.
3Reliability
If buffer drainage is performed over long periods to maintain sample integrity, then signal fidelity is preserved, but continuous monitoring of bat activity is prevented
Solution Approach 1:
The patent enables continuous monitoring by processing and outputting samples in continuous real-time frames rather than draining a complete buffer over long periods. The selective sample choosing process operates continuously on incoming frames, maintaining both signal fidelity through careful sample selection and monitoring continuity through immediate real-time output, thus resolving the contradiction between reliability and duration of action.
Data Source
AI summary
A method for listening to ultrasonic animal sound signals comprises: obtaining a plurality of input samples at an input sample rate, the plurality of input samples including at least one sequence of samples corresponding to an instance of an intermittently occurring animal sound signal; receiving a frame including at least two of the plurality of input samples; selecting a fraction of the samples as output samples, the fraction of the samples including a sample containing the at least one sequence of samples corresponding to an instance of an intermittently occurring animal sound signal; and transmitting the output samples at an output sample rate slower than the input sample rate.