基于非均匀B样条插值的EMD算法

EMD algorithm based on non-uniform B-spline interpolation

经验模态分解(EMD)是一种非平稳信号分析方法,存在分解不完全的问题。采用非均匀B样条插值生成包络线,得到了一种新的EMD算法,以实现准确的经验模态分解。针对同一信号给出的对比仿真,表明了算法的有效性。区别于传统算法的非均匀参数化方法具有精度高、实用性强的特点,利于多元推广。

Empirical mode decomposition（EMD） is a useful method in the nonlinear and non-stationary signal analysis.A novel EMD algorithm is proposed by non-uniform B-spline interpolation and solves the problem of incomplete signal decomposition effectively.Simulation results according to the same signal show that the new EMD algorithm is more suitable for the nonlinear and non-stationary signal.Different from the traditional algorithms,the non-uniform parameterization method has characteristics of high-precision and good-practical,which is conducive to promoting multi-dimensional.