Tikhonov Regularization for Multipath Interference Reduction in Mirror-Type Coastal Acoustic Tomographic Systems

This study presents a deconvolution method based on Tikhonov regularization to mitigate multipath interference in mirror-type coastal acoustic tomographic systems, thereby significantly improving the accuracy of ocean current estimation. The proposed approach is validated using data from a field experiment conducted in the Nekoseto Strait, Japan, in 2017. The results demonstrate effective extraction of arrival patterns under conditions of high channel temporal coherence <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$(&gt;0.9)$</tex-math></inline-formula>; however, performance degradation is observed during periods of rapid channel variability. The current speed estimates derived from the deconvolved mirror patterns show significantly closer alignment with reciprocal regular transmission estimates, compared to conventional methods that rely on first arrival picking. Our analysis reveals a strong positive correlation (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$R^{2} = 0.92$</tex-math></inline-formula>) between temporal coherence and extraction quality, underscoring the essential role of channel stability in the effectiveness of the deconvolution process.