Data Fusion of Multivariate Time Series Based on Local Weighted Zero-Order Prediction Algorithm

摘要

Utilizing multivariable data, the motion state of complex nonlinear systems could be well described by phase space reconstruction theory. In this paper, a novel fusion data algorithm is proposed via local weighted zero-order prediction algorithm. For effectively reserving the motion information of original system, the weighted coefficients are rationally designed Euclidean distance. Experimental results show that the algorithm is insensitive to the parameter embedding dimension m and delay time τ and can obtain desirable results on Lorenz model.