基于支持向量机的矿井无线信道建模与精确预测

发布时间：2018-06-17 20:50

本文选题：矿井巷道 + 电波传播　；参考：《西安科技大学》2015年硕士论文

【摘要】：煤炭在我国的经济发展中扮演着重要的角色,在一定时期内还无法被替代。煤矿开采过程中,通信系统为井下人员安全、提高生产效率以及生产过程优化提供重要保障。然而由于矿井巷道是一个特殊的受限空间,井下环境复杂多变,电磁波不能很好的传播,导致井下无线通信系统的发展相对较慢。本课题通过对井下电波传播特性分析研究,采用支持向量机建立矿井巷道无线信道模型,对井下无线通信系统的研究设计提供理论和实际意义。本文在对支持向量机、矿井巷道电磁波传播特性分析研究的基础上,以矩形巷道为对象,重点分析研究天线极化方式、载波频率、巷道围岩电参数、横截面尺寸、巷道壁粗糙程度、巷道壁倾斜、收发天线间距离等变化对电磁波的传输衰减的影响。针对大尺度衰落将上述影响因素作为支持向量机网络模型的输入,把电磁波的传输衰减作为模型的输出,以此训练支持向量机网络大尺度模型。在可行性分析阶段,实验数据通过计算机仿真获得,仿真模型选用了包括模式衰减、巷道壁粗糙度衰减和巷道倾斜衰减以及多径瑞利衰落在内的电波损耗叠加模型。针对于小尺度衰落信道,本文着重研究基于支持向量机算法的多径衰落信道预测模型,利用相空间重构理论对衰落信道的变化序列进行重构,并利用重构后的数据训练支持向量机信道衰落预测模型。文中支持向量机模型的核参数选择选用多层动态自适应优化算法。通过MATLAB仿真实验,基于支持向量机的矿井信道模型可以很好的学习巷道无线信道的变化,与其他传统模型比较,支持向量机信道模型的预测精度更高。实验结果证明,利用支持向量机技术建立井下电波传播模型是一个可行的途径。
[Abstract]:Coal plays an important role in China's economic development and can not be replaced in a certain period. In the process of coal mining, communication system provides an important guarantee for the safety of underground personnel, the improvement of production efficiency and the optimization of production process. However, due to mine roadway is a special restricted space, the underground environment is complex and changeable, electromagnetic wave can not spread well, resulting in the development of underground wireless communication system is relatively slow. In this paper, the radio channel model of mine roadway is established by using support vector machine, which provides theoretical and practical significance for the research and design of underground radio communication system. Based on the analysis of electromagnetic wave propagation characteristics of support vector machine and mine roadway, this paper focuses on the antenna polarization mode, carrier frequency, roadway surrounding rock electrical parameter, cross section size, taking rectangular roadway as the object. The influence of the roughness of tunnel wall, the inclination of tunnel wall and the distance between transmitters and transmitters on the attenuation of electromagnetic wave transmission. For large-scale fading, the above factors are taken as the input of the support vector machine network model, and the transmission attenuation of the electromagnetic wave is taken as the output of the model to train the large-scale support vector machine network model. In the stage of feasibility analysis, the experimental data are obtained by computer simulation, and the superposition model of radio wave loss including mode attenuation, roadway wall roughness attenuation, roadway slope attenuation and multipath Rayleigh fading is selected. For small scale fading channels, this paper focuses on the prediction model of multipath fading channels based on support vector machine (SVM) algorithm, and uses the theory of phase space reconstruction to reconstruct the variation sequences of fading channels. The reconstructed data is used to train the support vector machine channel fading prediction model. In this paper, the kernel parameter selection of support vector machine model is based on multi-layer dynamic adaptive optimization algorithm. Through MATLAB simulation experiment, the mine channel model based on support vector machine can learn the change of wireless channel of roadway well. Compared with other traditional models, the prediction accuracy of support vector machine channel model is higher. The experimental results show that it is a feasible way to establish downhole radio wave propagation model by using support vector machine (SVM) technology.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD655;TP18

【参考文献】