 |
|---|
| Architecture of the proposed deep neural network for data-driven direct localization. The model is comprised of three sub-models, which are initially trained individually to estimate range, azimuth and inclination. |
This page contains relevant materials for the data-driven direct localization project.
To download the supplementary material package of [1], click here.
To download the supplementary material package of [2], click here.
Related Papers
[1] Weiss, A., Arikan, T. and Wornell, G. W., “Direct Localization in Underwater Acoustics via Convolutional Neural Networks: A Data-Driven Approach”, in Proc. of IEEE Int. Workshop on Machine Learning for Signal Processing (MLSP), pp. 1–6, Aug. 2022. arXiv
[2] Weiss, A., Singer, A. C. and Wornell, G. W., “Towards Robust Data-Driven Underwater Acoustic Localization: A Deep CNN Solution with Performance Guarantees for Model Mismatch”, in Proc. of IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 1–5, May 2023. arXiv
[3] Weiss, A., Arikan, T., Vishnu, H., Deane, G. B., Singer, A. C. and Wornell, G. W., “A Semi-Blind Method for Localization of Underwater Acoustic Sources”, IEEE Trans. on Signal Processing, vol. 70, pp. 3090–3106, May 2022. arXiv