摘要
该文在离散小波变换理论和动态多尺度系统理论的基础上,建立了一种基于单传感器的多尺度状态融合估计新算法。该方法利用离散小波变换,对Kalman滤波模型的状态方程和观测方程分别进行多尺度处理,构建多尺度Kalman滤波模型,充分利用状态估计和观测数据在不同尺度上的特征进行融合估计,获得了优于单尺度Kalman滤波及已有多尺度状态融合估计方法的处理效果。并利用Monte Carlo仿真验证该算法的有效性。
On the basis of the theories of Discrete Wavelet Transform and Dynamic Multi-scale System, we proposed a novel algorithm for multi-scale fusion and estimation using single sensor in this paper. With discrete wavelet transform, we reformulated the state equation and observation equation of Kalman filter into a multi-scale form, in order to establish a no- vel multi-scale Kalman filtering model. By making full use of the signal feature on the diffident scales, the estimates ob-tained by use of the algorithm in this paper is more accurate than the results based on single scale Kalman filter and the multi-scale fusion estimating algorithm . A set of Monte Carlo simulation is performed, and the results show that our algo-rithm is effective and efficient as well.
出处
《信号处理》
CSCD
北大核心
2013年第8期971-976,共6页
Journal of Signal Processing
基金
国家科技支撑计划课题(2011BAH24B12)
国家自然科学基金委员会与中国民用航空局联合资助项目(61079008)
中国民航大学校内科研基金项目(09CAUC_E10)
中国民航大学科研启动基金(2011QD04S)资助
关键词
动态多尺度系统
离散小波变换
状态融合估计
卡尔曼滤波
Dynamic Multi-scale System
Discrete Wavelet Transform
Fusion and Estimation
Kalman Filter
