Song et al. (2025) Long-distance soil moisture monitoring via Helmholtz resonator–enhanced acoustic transmission and Swin-Transformer modeling

Identification

• Journal: Computers and Electronics in Agriculture
• Year: 2025
• Date: 2025-12-24
• Authors: Kangle Song, Jingbin Li, Yang Li, Jing Nie, Yuntao Sun, Wujun Zhang, Xiaojie Hou, Qiang Wang, Pengxiang Song
• DOI: 10.1016/j.compag.2025.111344

Research Groups

• College of Mechanical and Electrical Engineering, Shihezi University, Shihezi, China
• Key Laboratory of Northwest Agricultural Equipment, Ministry of Agriculture and Rural Affairs, Shihezi, China
• Xinjiang Production and Construction Corps Key Laboratory of Modern Agricultural Machinery, Shihezi, China
• Xinjiang Huier Zhilian Technology Co. Ltd., Changji, China
• Engineering Training Center, Xinjiang University, Urumqi, China

Short Summary

This study developed a Helmholtz resonator-enhanced acoustic transmission system and a Swin-Transformer model to overcome acoustic signal attenuation in soil, enabling long-distance and large-scale soil moisture monitoring with high accuracy.

Objective

• To design a resonance enhancement structure to overcome the short detection range of conventional acoustic methods and propose a large-scale transmission method for soil moisture detection based on a Helmholtz resonator.
• To construct a mapping model between acoustic time-frequency spectrograms and soil water content through feature extraction and model development.

Study Configuration

• Spatial Scale: Effective detection range extended from 4 meters to 60 meters; designed for large-scale and wide-area soil moisture monitoring.
• Temporal Scale: Implied continuous monitoring capability for agricultural water resource management.

Methodology and Data

• Models used: Swin-Transformer (for regression modeling of soil water content).
• Data sources: Acoustic signals transmitted through soil, enhanced by Helmholtz resonators; time-frequency spectrograms derived from these signals; real field conditions validation.

Main Results

• The Helmholtz resonator exhibited highly consistent response characteristics and strong frequency selectivity, significantly enhancing acoustic signal reception.
• The effective detection range of the system was extended from 4 meters to 60 meters.
• The Swin-Transformer-based regression model achieved a mean absolute error (MAE) of 0.207 %, a root mean square error (RMSE) of 0.244 %, and a coefficient of determination (R²) of 0.992 on the test dataset.
• In field trials, the model achieved an MAE of 1.046 %, an RMSE of 0.851 %, and an R² of 0.735.

Contributions

• Introduces a novel Helmholtz resonator-enhanced acoustic transmission method that significantly extends the effective detection range for soil moisture monitoring, addressing severe energy attenuation issues.
• Proposes a low-cost, transmission-based solution for wide-area soil moisture monitoring, offering a practical alternative to conventional methods.
• Develops and validates a Swin-Transformer-based regression model for accurate soil water content mapping from acoustic time-frequency spectrograms, including a visualization mechanism for interpretability.

Funding

• Not explicitly stated in the provided text.

Citation

@article{Song2025Longdistance,
  author = {Song, Kangle and Li, Jingbin and Li, Yang and Nie, Jing and Sun, Yuntao and Zhang, Wujun and Hou, Xiaojie and Wang, Qiang and Song, Pengxiang},
  title = {Long-distance soil moisture monitoring via Helmholtz resonator–enhanced acoustic transmission and Swin-Transformer modeling},
  journal = {Computers and Electronics in Agriculture},
  year = {2025},
  doi = {10.1016/j.compag.2025.111344},
  url = {https://doi.org/10.1016/j.compag.2025.111344}
}