Local propagation speed constrained estimation of the slowness vector from non-planar array observations

Area of Science:

• Geophysics
• Acoustics
• Signal Processing

Background:

• Infrasound wave slowness vector estimation is crucial for determining parameters like direction of arrival.
• Topographical variations in sensor altitudes introduce bias in traditional horizontal plane estimations.
• Unbiased 3D estimation reduces bias but increases variance, impacting performance.

Purpose of the Study:

• To investigate the benefits of incorporating a priori constraints on the slowness vector for 3D infrasound estimation.
• To thoroughly analyze the bias and variance of a constrained 3D estimator, addressing a gap in existing literature.
• To propose a constrained 3D estimator offering an optimal bias/variance trade-off.

Main Methods:

• Development of a constrained 3D estimator for infrasound wave slowness vectors.
• Theoretical computation of bias and variance using an asymptotic Gaussian approximation.
• Validation through simulations using both synthetic and real infrasound data.

Main Results:

• The constrained 3D estimator significantly reduces variance compared to unbiased methods.
• The study quantifies the bias and variance trade-offs associated with the constrained estimator.
• Simulations confirm the theoretical findings for both synthetic and real-world data.

Conclusions:

• Constraining the slowness vector in 3D infrasound estimation provides a beneficial bias-variance compromise.
• This approach enhances estimation performance, particularly when accurate wave speed knowledge is available.
• The proposed constrained 3D estimator offers improved accuracy for infrasound array analysis.