非最小相位FIR系统的自适应辨识
Adaptive Identification of Nonminimum Phase FIR Systems
-
摘要: 提出了一种基于线性代数方程组约束和梯度法的非最小相位FIR系统的自适应辨 识算法.选用的线性方程组具有列满秩的系数矩阵,保证了系统参数的唯一可识别性.由于只 采用高阶累量,故能够抑制任何高斯有色噪声的影响.重点讨论了梯度法中步长的选择,提出 了收敛速度最快的变步长.仿真实验的结果证实了算法的有效性.Abstract: Based on linear equation constraints and steepest descent method, an adaptive algorithm is proposed for nonminimum phase FIR system identification. The uniqueress of identification is guaranteed by the full rank of the coefficient matrix of the utilized linear equations. Effects of colored Gaussian noise are automatically handled because of using merely higher-order cumulants. The optimal variable step sizes are discussed in the sense of convergence rate. Effectiveness of the new algorithm is verified by simulation results.
-
计量
- 文章访问数: 2401
- HTML全文浏览量: 109
- PDF下载量: 1000
- 被引次数: 0
下载:
