Joint Source-Environment Adaptation for Deep Learning-Based Underwater Acoustic Source Ranging

2025-03-30

AI Summary

This paper addresses the poor generalization of pretrained deep models for underwater acoustic source ranging in novel oceanic environments. We propose an unsupervised cross-environment adaptation method that requires neither target-domain labels nor access to original training data. Our approach innovatively couples unsupervised domain adaptation (UDA) with joint source–environment modeling based on received signal energy, overcoming the limitation of conventional single-environment dependency. The method integrates Bellhop acoustic field simulation, KAM11-measured oceanic noise injection, and energy-feature-driven neural network fine-tuning. Evaluated in a SWellEx-96-like simulated environment, it reduces mean ranging error by 32% over baseline methods and significantly enhances robustness to realistic oceanic noise. This work establishes a new paradigm for unsupervised domain adaptation in underwater acoustics.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments

Abstract

In this paper, we propose a method to adapt a pre-trained deep-learning-based model for underwater acoustic localization to a new environment. We use unsupervised domain adaptation to improve the generalization performance of the model, i.e., using an unsupervised loss, fine-tune the pre-trained network parameters without access to any labels of the target environment or any data used to pre-train the model. This method improves the pre-trained model prediction by coupling that with an almost independent estimation based on the received signal energy (that depends on the source). We show the effectiveness of this approach on Bellhop generated data in an environment similar to that of the SWellEx-96 experiment contaminated with real ocean noise from the KAM11 experiment.

Problem

• Adapting pre-trained model to new underwater environments
• Improving generalization via unsupervised domain adaptation
• Enhancing ranging accuracy using signal energy estimation

Innovation

• Unsupervised domain adaptation for model generalization
• Fine-tuning pre-trained network without target labels
• Coupling model prediction with signal energy estimation