一种基于支持向量机的电波传播损耗计算方法  

A Method of Radio Wave Propagation Loss Calculation Based on Machine Learning

作　　者：李炳呈 肖逸凡 李佳霖 LI Bingcheng;XIAO Yifan;LI Jialin(College of Communication Engineering,Army Engineering University of PLA,Nanjing 210007,China)

机构地区：[1]陆军工程大学通信工程学院,江苏南京210007

出　　处：《电声技术》2022年第5期130-133,共4页Audio Engineering

摘　　要：当前,电磁环境日益复杂,电波传播损耗作为电磁环境分析领域的重要研究对象,其预测结果的准确性将直接影响频谱态势生成、用频兼容分析等实际效果。由于电磁波在传播过程中易受环境影响,传统方法采用单一模型计算电波传播损耗,潜在误差较大。本文提出一种基于支持向量机(Support Vector Machine,SVM)的电波传播损耗计算方法。在训练阶段,采用不同传播环境下的感知数据对支持向量机模型进行训练;在测试阶段,以一定步长对传播路径进行分段,运用训练得到的支持向量机模型来对传播环境分段识别,继而采用各段识别结果相匹配的电波传播损耗模型分段计算损耗值并叠加为总的路径损耗值。仿真结果表明,相较于采用单一模型的传统方法,本文所提方法能够显著提高电波传播损耗计算精度。Nowadays, the electromagnetic environment is becoming more complicated. As an important research subject in the field of electromagnetic environment analysis, the accuracy of radio wave propagation loss will directly influence the actual effects of spectrum situation construction and frequency compatibility analysis. Given that the electromagnetic wave is easily affected by the environment in the process of propagation, the traditional method uses a single model to calculate the radio wave propagation loss, which has a large potential error. In this paper, a calculation method of radio wave propagation loss based on support vector machine is proposed. In the training phase, the Support Vector Machine(SVM) model is trained on the sensed data in different propagation environments;In the test stage, the propagation path is splited with a certain step size, and the trained support vector machine model is used to identify the propagation environment by segments. Then, the radio wave propagation loss model matching the recognition results of all segments is used to calculate the loss values and superimpose them into the total path loss value. The simulation results show that compared with the traditional method using a single model, the proposed method can significantly improve the calculation accuracy of radio wave propagation loss.

关 键 词：电波传播损耗 支持向量机 分段识别 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]