Local Prediction of Chaotic Time Series Based on Support Vector Machine

Based on phase space delay-coordinate reconstruction of a chaotic dynamics system, we propose a local prediction of chaotic time series using a support vector machine （SVM） to overcome the shortcomings of traditional local prediction methods. The simulation results show that the performance of this proposed predictor for making onestep and multi-step prediction is superior to that of the traditional local linear prediction method and global SVM method. In addition, it is significant that its prediction performance is insensitive to the selection of embedding dimension and the number of nearest neighbours, so the satisfying results can be achieved even if we do not know the optimal embedding dimension and how to select the number of nearest neighbours.     

时空混沌序列的局域支持向量机预测

结合局域预测法计算速度快的优点和支持向量机的泛化性能好、全局最优、稀疏解等特性，用局域支持向量机预测研究了时空混沌序列的局域预测性能，并用局域支持向量机预测模型讨论了嵌入维数、邻近个数选择以及时空混沌的耦合方式和格子间的耦合强度变化对时空混沌局域预测性能的影响.研究结果表明：局域支持向量机不仅比全局支持向量机、局域零阶预测、局域线性预测等方法具有更好的预测性能，且具有对嵌入维数和邻近个数不敏感的优点；时空混沌的耦合方式和格子间的耦合强度对时空混沌序列的预测性能有明显影响.     

连续混沌信号的离散余弦变换域二次实时滤波预测

提出了少参数二阶Volterra滤波器的一种离散余弦变换(DCT)域二次滤波实现结构及其NLMS自适应算法，并用这种DCT域二次滤波预测器研究了三种连续混沌信号的非线性实时多步预测性能. 仿真研究结果表明：(1) 这种DCT域二次滤波预测器比少参数二阶Volterra滤波器的一步预测均方误差性能提高了100倍，表明这种实现结构简单、易实现，且具有更好的收敛性能；（2）采用这种滤波预测器对三种连续混沌时间序列的实时多步预测性能明显优于局域法的多步预测性能. 关键词： 混沌 实时预测 NLMS自适应算法     

混沌时间序列的一种局域离散余弦变换域Volterra预测

提出了一种局域离散余弦变换（DCT）域Volterra预测，并用于混沌时间序列预测。DCT被用来减少Volterra预测器的矩阵计算复杂性。数值仿真结果表明：本文提出的方法比传统的局域线性预测方法能更有效地预测混沌时间序列和预测精度。     

Neural Volterra filter for chaotic time series prediction

A new second-order neural Volterra filter （SONVF） with conjugate gradient （CG） algorithm is proposed to predict chaotic time series based on phase space delay-coordinate reconstruction of chaotic dynamics system in this paper, where the neuron activation functions are introduced to constraint Volterra series terms for improving the nonlinear approximation of second-order Volterra filter （SOVF）. The SONVF with CG algorithm improves the accuracy of prediction without increasing the computation complexity. Meanwhile, the difficulty of neuron number determination does not exist here. Experimental results show that the proposed filter can predict chaotic time series effectively, and one-step and multi-step prediction performances are obviously superior to those of SOVF, which demonstrate that the proposed SONVF is feasible and effective.