Time of arrival difference estimation for narrow band high frequency echolocation clicks.

Area of Science:

• Bioacoustics
• Marine Mammal Research
• Signal Processing

Background:

• Harbor Porpoise echolocation clicks are high-frequency signals used for navigation and foraging.
• Accurate estimation of time of arrival differences (TDOA) is essential for localization.
• Spatial aliasing with closely spaced hydrophones can introduce significant errors in TDOA measurements.

Purpose of the Study:

• To develop and present algorithms for accurate TDOA estimation of Harbor Porpoise echolocation clicks.
• To investigate the impact of spatial aliasing on TDOA accuracy at different sampling rates.
• To propose methods for mitigating errors in TDOA estimation.

Main Methods:

• Development of algorithms for cross-correlation analysis of echolocation click signals.
• Simulation and analysis of TDOA errors under conditions of spatial aliasing.
• Implementation of data up-sampling and parabolic interpolation techniques.
• Evaluation of error reduction for signals with signal-to-noise ratio (SNR) > 10 dB.

Main Results:

• Spatial aliasing can lead to incorrect peak selection in cross-correlation functions, causing large TDOA errors.
• At sampling rates less than approximately 6 times the fundamental frequency, incorrect peak selection occurs in 55% of measurements.
• Up-sampling and parabolic interpolation significantly reduce TDOA errors by over two orders of magnitude for clicks with SNR > 10 dB.

Conclusions:

• The presented algorithms effectively mitigate TDOA errors in Harbor Porpoise echolocation data caused by spatial aliasing.
• Accurate TDOA estimation is achievable even with closely spaced hydrophones through advanced signal processing techniques.
• These findings enhance the reliability of passive acoustic monitoring for marine mammal research.