摘要
本文基于多尺度卡尔曼滤波方法来估计淹没在加性高斯白噪声中的分形布朗运动.针对每一 尺度，给出了相应的动态系统参数和运动模型方程以及更精确的估计算法.并与多尺度维纳 滤波进行了对比，计算机仿真结果证明了其优越性.
A filter bank design based on orthonormal wavelets and equipped with a multiscale Kalman filter was recently proposed for estimating fractal Brownian motion in additive Gaussian white noise.In this paper,we give the corresponding parameters of the dynamic system and more accurate estimation algorithm.Comparisons between Wiener and Kalman filters are given.Typical computer simulation results demons trate its feasibility and effectiveness.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2001年第9期1157-1160,共4页
Acta Electronica Sinica
基金
国家自然科学基金项目(No.19971063)
国家自然科学基金重点项目(No.69732010 )
关键词
分形随机信号
卡尔曼滤波
小波变换
信号处理
fractal stochastic signal
fractional Brownian motion
1/f processes
Kalman filter ing
wavelet transform
