Exploring Transformer-Based Direction-of-Arrival Estimation Over Sea Surface: A BERT Approach With Physics-Based Loss Function
- 7 August 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-13
Abstract
A comprehensive exploration of the transformer and dual receiver system-based direction-of-arrival (DOA) estimation is presented in the context of sea surface scattering, particularly under varying sea conditions. A bidirectional encoder representation from the transformer (BERT) with a physics-based loss function is utilized to process two individual channel radars. The datasets are the radar scattering coefficients of sea surface simulated at C-band for co- and cross polarizations. Through detailed analysis of simulated datasets and Root- Mean-Squared Error (RMSE) evaluations, the model’s performance is investigated across different observation modes, namely the co-polarization (CP), the co-azimuth (CA) and full-bistatic (FB), and Beaufort wind scale from 3 to 5. Our study demonstrates that the bidirectional encoder representation from transformer model, employing a physics-based loss function, outperforms the baseline long short-term memory (LSTM) model, especially under high noise levels and with larger datasets. Significant correlations between wind conditions and DOA accuracy are observed, highlighting the bidirectional encoder representation from transformer model’s adaptability to dynamic environmental factors, particularly under increased wind scales. The choice of observation mode, with CP and FB consistently outperforming CA, proves pivotal. Precise simulation of speckle variations and optimized observation mode selection are identified as crucial avenues for enhancing the model’s practical utility.Keywords
Funding Information
- China Scholarship Council, National Natural Science Foundation of China, (42201352)
- Natural Science Foundation of Jiangsu Province (BK20231459)
- Nanjing University Initiative Research
This publication has 46 references indexed in Scilit:
- Study on direction dependence of the fully polarimetric wind-induced ocean emissivity at L-band using a semi-theoretical approach for Aquarius and SMAP observationsRemote Sensing of Environment, 2021
- Self-attention for raw optical Satellite Time Series ClassificationISPRS Journal of Photogrammetry and Remote Sensing, 2020
- Spaceborne bi- and multistatic SAR: potential and challengesIEE Proceedings - Radar, Sonar and Navigation, 2006
- Array self-calibration with large sensor position errorsSignal Processing, 2001
- Array shape calibration using eigenstructure methodsSignal Processing, 1991
- Maximum likelihood methods for direction-of-arrival estimationIEEE Transactions on Acoustics, Speech, and Signal Processing, 1990
- Novel eigenanalysis method for direction estimationIEE Proceedings F Radar and Signal Processing, 1990
- ESPRIT-estimation of signal parameters via rotational invariance techniquesIEEE Transactions on Acoustics, Speech, and Signal Processing, 1989
- On the statistics of K-distributed noiseJournal of Physics A: General Physics, 1980
- Reflection of electromagnetic waves from slightly rough surfacesCommunications on Pure and Applied Mathematics, 1951