【作者】 甘玉涛；

【导师】 王振朝；

【作者基本信息】 河北大学 ， 通信与信息系统， 2009， 硕士

【摘要】 低压PLC信道是一个受某些确定性规律控制的复杂非线性有限维动力系统,研究其特性并建立信道模型,是低压PLC技术领域的重要课题。本研究将混沌理论引入低压PLC信道特性研究,本研究的主要工作及成果如下：一、利用自制数据采集系统,将50Hz的工频信号滤除,利用示波器从低压PLC信道直接进行数据采集。二、通过定量计算低压PLC信道实测时间序列的饱和关联维数和最大Lyapunov指数等不变混沌特征量,证明了低压PLC信道具有混沌特性。三、利用支持向量机对信道实测混沌时间序列进行预测,并对支持向量机各参数对预测性能的影响进行了研究,总结了各参数间的关系及预测的具体实现步骤。四、计算低压PLC信道实测时间序列的嵌入维数和延迟时间,对时间序列进行相空间重构。在重构的相空间中,利用遗传编程算法对实测时间序列进行了建模。更多还原

【Abstract】 The low-voltage power line communication (L-PLC) channel is a complicated nonlinear finite-dimensional dynamical system and it is controlled by some determinative regulations. It is important to research the characteristics of L-PLC channel and estabilish the channel model for L-PLC technology. In this paper, chaos theory is applied to research the characteristics of L-PLC channel. The main works and research results of this paper are following:1. A lot of experimental data are collected by data measurement system. The frequency signal isolated and the time series are measured from lower power line by oscilloscopes directly.2. The chaotic characteristics value of experimental time series are calculated. The result proves that the timse series measured from L-PLC channel are weak chaotic. Therefore the channel is chaotic, too.3. Support Vector Machines (SVM) is used to predict the chaotic time series of the channel of L-PLC. And then the predition of the characteristics of L-PLC cahnnel is realized. The influences of the parameters of SVM on the predition are also discussed. The relationships between the parameters and the specific steps of prediction are summarize.4. The embedding dimensions and time delays are calculated and then the phase space is reconstructed. In the reconstructed phase space, Genetic Programming(GP) with strong global analytical ability is used in modeling of measured time series.更多还原