Scholarlink

Opportunistic sensing through Satellite Microwave Links (SMLs) offers a cost-effective and scalable solution for real-time rainfall monitoring. In this study, Smart Rainfall System (SRS) data from three links were used to classify wet/dry periods using two machine learning models: Random Forest (RF) and a Fully Connected (FC) neural network. Classification was conducted at four time windows (15, 30, 60, and 120 min) to assess the influence of window size on feature extraction and model performance. Results show that both models effectively detect rainfall, with shorter windows enhancing classification accuracy.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Multi-time-window Machine Learning Approach for Rainfall Detection

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mostafa Ftouni,
  • Christian Gianoglio,
  • Margherita Spalla,
  • Matteo Colli,
  • Andrea Randazzo

摘要

Opportunistic sensing through Satellite Microwave Links (SMLs) offers a cost-effective and scalable solution for real-time rainfall monitoring. In this study, Smart Rainfall System (SRS) data from three links were used to classify wet/dry periods using two machine learning models: Random Forest (RF) and a Fully Connected (FC) neural network. Classification was conducted at four time windows (15, 30, 60, and 120 min) to assess the influence of window size on feature extraction and model performance. Results show that both models effectively detect rainfall, with shorter windows enhancing classification accuracy.