Objective detection and time-frequency localization of components within transient signals

Area of Science:

• Acoustics
• Signal Processing
• Bioacoustics

Background:

• Analyzing overlapping transient pulses in multi-component signals is challenging.
• Accurate detection of Gaussian shaped transients is crucial in various acoustic research fields.
• Existing methods may struggle with closely located pulses and disruptive noise.

Purpose of the Study:

• To present and evaluate an automatic component detection method for overlapping transient pulses.
• To demonstrate the effectiveness of the scaled reassignment technique for transient pulse analysis.
• To provide a robust tool for analyzing coinciding Gaussian shaped transients.

Main Methods:

• An automatic component detection method was developed.
• The scaled reassignment technique was employed for high-resolution analysis.
• The method was tested on laboratory pulse-echo data and dolphin echolocation signals.

Main Results:

• The scaled reassignment technique achieved superior resolution for closely located Gaussian shaped transient pulses.
• The method automatically detected and counted transients with accurate center times and frequencies.
• Effective performance was observed even in the presence of heavy disruptive noise.

Conclusions:

• The presented automatic method offers high accuracy in detecting and characterizing overlapping transient pulses.
• The technique shows significant potential for applications in acoustic research, particularly in analyzing coinciding Gaussian shaped transients.
• The method's minimal user input requirement facilitates its easy adoption in diverse research projects.