Performance comparisons of matched field processing and machine learning-based source range estimation under mismatch conditions

Soyeon Park; Geunhwan Kim; Gihoon Byun

doi:10.7776/ASK.2025.44.4.392

All Issue

2025 Vol.44, Issue 4 Preview Page Next Page

Research Article

Performance comparisons of matched field processing and machine learning-based source range estimation under mismatch conditions 오정합 조건에 따른 정합장 처리와 머신러닝 기반 음원 거리 추정 성능 비교

31 July 2025. pp. 392-406

PDF XML

Abstract

Matched Field Processing (MFP) heavily depends on the accuracy of environmental information, and its localization performance significantly degrades when model mismatch occurs. To overcome this limitation, a study on machine learning-based matched field processing, which estimates source range without an acoustic model, was proposed. However, a systematic analysis on mismatch problems has not been conducted. In this study, using real ocean experimental data and numerical simulations, we compare and analyze the performance of model-based MFP and machine learning-based MFP under representative mismatch conditions such as bathymetry mismatch and array tilt mismatch. As a result, both methods exhibited deviations in range estimation due to mirage effects caused by bathymetric mismatch, and the estimation error increased sharply when the array tilt mismatch exceeded 1.3°. In addition, performance degradation was more severe under combined mismatch conditions than under single mismatch. These results show that the performance of machine learning-based MFP can also vary significantly depending on mismatch conditions.

Keywords

Source-range estimation

Matched Field Processing (MFP)

Machine learning

Mismatch

정합장 처리(Matched Field Processing, MFP)는 환경 정보의 정확도에 크게 의존하며 모델 오정합이 발생할 경우 추정 성능이 현저히 저하된다. 이러한 한계를 극복하기 위해, 음향 모델 없이 음원의 거리를 추정하는 머신러닝 기반 정합장 처리에 대한 연구가 제안되었으나, 오정합 문제에 대한 체계적인 분석은 이루어지지 않았다. 본 연구에서는 실제 해상 실험 데이터와 수치 시뮬레이션을 활용하여, 수심 불일치 및 수신배열 센서 기울기 불일치와 같은 대표적인 오정합 조건에서 음향 모델 기반 정합장 처리와 머신러닝 기반 정합장 처리의 성능을 비교 분석하였다. 분석 결과, 두 기법 모두 수심 오정합 시 신기루 현상으로 인해 거리 추정에 편차가 발생하였으며, 배열 기울기 불일치가 1.3°를 초과할 경우 오차가 급격히 증가하였다. 또한, 단일 오정합 조건보다 복합 오정합 조건에서 성능 저하가 더욱 심화되는 경향을 보였다. 이러한 결과는 머신러닝 기반 정합장 처리 역시 오정합 조건에 따라 거리 추정 성능이 크게 달라질 수 있음을 보여준다.

키워드

거리 추정

정합장 처리

머신러닝

오정합

References

A. B. Baggeroer, W. A. Kuperman, and P. N. Mikhalevsky, “An overview of matched field methods in ocean acoustics,” IEEE. J. Oceanic Eng. 18, 401-424 (1993).

10.1109/48.262292

A. Tolstoy, Matched Field Processing for Underwater Acoustics (World Scientific, Singapore, 1993), pp. 1-228.

Z. H. Michalopoulou and M. B. Porter, “Matched-field processing for broadband source localization,” IEEE J. Ocean Eng. 21, 384-392 (1996).

10.1109/48.544049

G. J. Orris, M. Nicholas, and J. S. Perkins, “The matched-phase coherent multi-frequency matched field processor,” J. Acoust. Soc. Am. 107, 2563-2575 (2000).

10.1121/1.428644

C. Soares and S. M. Jesus, “Broadband matched-field processing: Coherent and incoherent approaches,” J. Acoust. Soc. Am. 113, 2587-2598 (2003).

10.1121/1.1564016

C. Debever and W. A. Kuperman, “Robust matched-field processing using a coherent broadband white noise constraint processor,” J. Acoust. Soc. Am. 122, 1979-1986 (2007).

10.1121/1.2769830

H. P. Bucker, “Use of calculated sound fields and matched-field detection to locate sound sources in shallow water,” J. Acoust. Soc. Am. 59, 368-373 (1976).

10.1121/1.380872

G. L. D’Spain, J. J. Murray, W. S. Hodgkiss, N. O. Booth, and P. W. Schey, “Mirages in shallow water matched field processing,” J. Acoust. Soc. Am. 105, 3245-3265 (1999).

10.1121/1.424653

G. Byun, C. Cho, H. C. Song, J. S. Kim, and S. H. Byun, “Array invariant-based calibration of array tilt using a source of opportunity,” J. Acoust. Soc. Am. 143, 1318-1325 (2018).

10.1121/1.5025844

G. Byun, F. H. Akins, K. L. Gemba, H. C. Song, and W. A. Kuperman, “Multiple constraint matched field processing tolerant to array tilt mismatch,” J. Acoust. Soc. Am. 147, 1231-1238 (2020).

10.1121/10.0000784

J-S. Park, J. Park, S-U. Son, H. Bae, and K. Lee, “Characteristics of source localization with horizontal line array using frequency-difference autoproduct in the East Sea environment,” (in Korean), J. Acoust. Soc. Kr. 43, 29-38 (2023).

H. Niu, E. Reeves, and P. Gerstoft, “Source localization in an ocean waveguide using supervised machine learning,” J. Acoust. Soc. Am. 142, 1176-1188 (2017).

10.1121/1.5000165

H. Niu, E. Ozanich, and P. Gerstoft, “Ship localization in Santa Barbara Channel using machine learning classifiers,” J. Acoust. Soc. Am. 142, EL455-EL460 (2017).

10.1121/1.5010064

H. C. Song, C. Cho, W. S. Hodgkiss, S. Nam, S. Kim, and B. Kim, “Underwater sound channel in the northeastern East China Sea,” Ocean Eng. 147, 370-374 (2018).

10.1016/j.oceaneng.2017.10.045

M. B. Poter, The Acoustics Toolbox, http://oalib.hlsresearch.com/AcousticsToolbox/ (Last viewed 5 May 2025).

G. Byun, H. C. Song, and J. S. Kim, “Performance comparisons of array invariant and matched field processing using broadband ship noise and a tilted vertical array,” J. Acoust. Soc. Am. 144, 3067-3074 (2018).

10.1121/1.5080603

B. M. Worthmann, H. C. Song, and D. R. Dowling, “High frequency source localization in a shallow ocean sound channel using frequency-difference matched field processing,” J. Acoust. Soc. Am. 138, 3549-3562 (2015).

10.1121/1.4936856

Information

Publisher :The Acoustical Society of Korea
Publisher(Ko) :한국음향학회
Journal Title :The Journal of the Acoustical Society of Korea
Journal Title(Ko) :한국음향학회지
Volume : 44
No :4
Pages :392-406
Received Date : 2025-05-13
Revised Date : 2025-06-20
Accepted Date : 2025-07-07
DOI :https://doi.org/10.7776/ASK.2025.44.4.392

The Journal of the Acoustical Society of KoreaISSN:1225-4428(Print) 2287-3775(Online)한국음향학회

All Issue