All Issue

2025 Vol.44, Issue 3 Preview Page

Research Article

A biomimetic model for target identification using active sonar

능동 소나 기반 표적 식별을 위한 생체모방모델

Sangwook Park1*
박 상욱1*
1강릉원주대학교 전자반도체공학부
*Corresponding Author
31 May 2025. pp. 223-230
PDF XML Citation
Abstract

This study proposes BioNetV2, an artificial neural network inspired by the biosonar of bats, to improve the echo signal representation in noisy and cluttered environments. The conventional BioNet suffers from low time resolution and limited training data, leading to decreased target classification performance and a sharp decline in discriminability under noisy conditions. To address these limitations, BioNetV2 incorporates high-resolution temporal features and learns from a more diverse set of echo signals, enhancing its performance. In the experiments, echo data were collected from six different objects and three reflection angles, and the discriminability of 13 target classes was evaluated using Fisher distance in both noisy and cluttered environments. The results demonstrate that BioNetV2 outperforms conventional methods (BioNet, AudSpec, and CA_Gabor) in discriminability, with the least performance degradation observed in the Signal to Noise Ratio (SNR) –10 dB condition, proving its robustness in challenging environments. Future research will focus on integrating sequential echo observations to extend the model toward 3D object recognition.

Keywords
Biomimetic model
Active sonar
Target identification
Autoencoder

본 연구는 박쥐의 생체 소나(Sound Navigation and Ranging, SONAR)를 모방한 인공신경망 모델(BioNetV2)을 제안하여, 잡음 및 클러터 환경에서의 반사음 표현 성능을 개선한다. 기존 BioNet은 낮은 시간 해상도와 제한된 학습 데이터로 인해 표적 식별 성능이 저하되며, 잡음환경에서 분별력이 급격히 감소하는 한계를 가진다. 이를 해결하기 위해, BioNetV2는 고해상도 시간 해상도를 적용하고, 다양한 반사음 데이터를 학습하여 성능을 향상시켰다. 실험에서는 6가지 물체와 3가지 반사면을 고려하여 반사음 데이터를 수집하고, 잡음 및 클러터 환경에서 Fisher distance를 기반으로 13개 클래스의 분별력을 평가하였다. 실험 결과, BioNetV2는 기존 방법(BioNet, AudSpec, CA_Gabor)보다 높은 분별력을 보였으며, 특히 신호 대 잡음비(Signal to Noise Ratio, SNR) –10 dB 환경에서도 성능 저하가 가장 적어 강인한 반사음 표현 능력을 입증하였다. 향후 연구에서는 연속적인 반사음 정보를 통합하여 3D 객체 인식으로 확장할 계획이다.

키워드
생체모방모델
능동 소나
표적 식별
오토인코더
References
1

D. P. Williams, F. Baralli, M. Micheli, and S. Vasoli, "Adaptive underwater sonar surveys in the presence of strong currents," Proc. ICRA, 2604-2611 (2016).

10.1109/ICRA.2016.7487418
2

M. Franchi, A. Ridolfi, and B. Allotta, "Underwater navigation with 2D forward looking SONAR: Adaptive unscented Kalman filter based strategy for AUVs," J. Field. Robot. 38, 355-385 (2020).

10.1002/rob.21991
3

A. Kumar, R. Kumar, M. Chandra, and K. Kishore, "Study of under-water sonar system for change in propagation speed, depth of water, bottom loss and estimating optimal PDFs," Proc. 6th ISCON, 1-6 (2023).

10.1109/ISCON57294.2023.10112071
4

Y. Tang, L. Wang, S. Jin, J. Zhao, C. Huang, and Y. Yu, "AUV-based side-scan sonar realtime method for underwater-target detection," J. Mar. Sci. Eng. 11, 690 (2023).

10.3390/jmse11040690
5

I. E. Dror, M. Zagaeski, and C. F. Moss, "Three- dimensional target recognition via sonar: A neural network model," Neural Networks, 8, 149-160 (1995).

10.1016/0893-6080(94)00057-S
6

J. M. Fialkowski and R. C. Gauss, "Methods for identifying and controlling sonar clutter," IEEE J. Ocean. Eng. 35, 330-354 (2010).

10.1109/JOE.2010.2044912
7

H. Cho, J. Gu, and S.-C. Yu, "Robust sonar-based underwater object recognition against angle-of-view variation," IEEE Sens. J. 16, 1013-1025 (2016).

10.1109/JSEN.2015.2496945
8

S. Andoni, N. Li, and G. D. Pollak, "Spectrotemporal receptive fields in the inferior colliculus revealing selectivity for spectral motion in conspecific vocalizations," J. Neurosci. 27, 4882-4893 (2007).

10.1523/JNEUROSCI.4342-06.200717475796PMC6672083
9

S. Park, A. Salles, K. Allen, C. F. Moss, and M. Elhilali, "Natural statistics as inference principles of auditory tuning in biological and artificial midbrain networks," eNeuro, 8, ENEURO.0525-20.2021 (2021).

10.1523/ENEURO.0525-20.202133947687PMC8211468
10

S. Park, A. Salles, K. Allen, C. Moss, and M. Elhilali, "Biomimetic mappings for active sonar object recognition in clutter," Proc. ICASSP, 6265-6269 (2024).

10.1109/ICASSP48485.2024.10446338
11

T. Chi, P. Ru, and S. A. Shamma, "Multiresolution spectrotemporal analysis of complex sounds," J. Acoust. Soc. Am. 118, 887-906 (2005).

10.1121/1.194580716158645
12

NSL Toolbox, https://github.com/tel/NSLtools/tree/master, (Last viewed April 11, 2025).

13

M. Elhilali, A. S, Shamma, Z. J. Simon, and B. J. Fritz, "A linear systems view to the concept of STRF," in Handbook of Modern Techniques in Auditory Cortex (Otolaryngology Research Advances), edited by D. A. Depireux and E. Mounya (Nova Science, New York, 2013).

14

A. Salles, S. Park, H. Sundar, S. Macías, M. Elhilali, and C. F. Moss, "Neural response selectivity to natural sounds in the bat midbrain," Neuroscience, 434, 200-211 (2020).

10.1016/j.neuroscience.2019.11.04731918008
15

K. M. Allen, A. Salles, S. Park, M. Elhilali, and C. F. Moss, "Effect of background clutter on neural discrimination in the bat auditory midbrain," J. Neurophysiol. 126, 1772-1782 (2021).

10.1152/jn.00109.202134669503PMC8794058
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 :3
  • Pages :223-230
  • Received Date : 2025-02-27
  • Revised Date : 2025-03-17
  • Accepted Date : 2025-03-21
  • DOI :https://doi.org/10.7776/ASK.2025.44.3.223
Journal Informaiton The Journal of the Acoustical Society of Korea The Journal of the Acoustical Society of Korea
E-Book
  • scopus_JASK
  • ESCI_JASK
  • Google scholar_JASK
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • ccl
Journal Informaiton Journal Informaiton - close