摘要
功率谱估计中,解析公式法由于未能解决白噪声与系统输出之间的相关性问题,计算结果误差较大且稳定性较差,为此在谱估计中引入Kalm an滤波.对于某一给定的AR过程来设计Kalm an滤波器,在滤波器的迭代运算中把AR过程的激励和AR过程的系统输出作为滤波器的输入、白噪声作为滤波器参数,并将该滤波器融合于谱估计的计算当中.仿真结果表明,与解析公式相比,Kalman滤波计算精度和结果稳定性都有较大提高,可以作为解决此类问题的选择之一.
In power spectrum estimation,the formula analysis method can result in large errors and less stability due to correlation between white noise and system response.Kalman filtering was introduced to deal with this problem.For a given autoregressive（AR） process,a Kalman filter must be designed.In the iterative operating process of the Kalman filter,the impulse of the AR process and the system output of the AR process are viewed as the input vector of the Kalman filter,white noise as parameters,and Kalman filter is combined into a power spectrum estimation of the AR process.Simulation results show that computational precision is improved and stability is enhanced,which proves that the Kalman filter is an effective alternative tool for power spectrum estimation in AR processes.
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
北大核心
2010年第1期115-119,共5页
Journal of Harbin Engineering University
基金
"十一五"国家科技支撑计划重点项目(2009BADB9B08)
