通信与信息技术Issue(2)：36-40,5.
一种机器学习协同卫星图像的电波传播模型
A machine learning-based radio wave propagation model for collaborative satellite images
邵梓铭 1罗业超 1李晋 1邵尉 1刘杨1
作者信息
- 1. 中国人民解放军陆军工程大学,江苏 南京 210007
- 折叠
摘要
Abstract
With the rapid development of wireless communication technology and the acceleration of urbanization,the study of radio wave propagation characteristics in urban environments has become a core subject in modern communication fields.Our research propos-es an ensemble model based on Convolutional Neural Network and Random Forest to predict the effect of the radio wave propagation.This model constructs a multimodal dataset with structured data and satellite images,and devises an efficient ensemble learning by combining the feature extraction of CNN for image data and the model fitting of Random Forest for structured data,which can better capture the com-plex features of urban environments and improve the prediction performance of the model.Experimental results indicate that the proposed ensemble model achieves RMSE and R2 of 2.615dB and 0.924 respectively on our test-set,significantly outperforming the single models and the baseline models,verifying its effectiveness and superiority.关键词
卷积神经网络/随机森林/集成模型/卫星图像/多模态数据集Key words
Convolutional neural network/Random forest/Ensemble model/Satellite images/Multimodal dataset分类
信息技术与安全科学引用本文复制引用
邵梓铭,罗业超,李晋,邵尉,刘杨..一种机器学习协同卫星图像的电波传播模型[J].通信与信息技术,2026,(2):36-40,5.
