1. Nonlinear frequency domain spline prioritization optimization generalized maximum correntropy criterion evolved momentum adaptive filtering.
- Author
-
Chen, Xixian and Liu, Zhen
- Abstract
The interference of impulsive noise is very common in the identification of nonlinear systems. The spline prioritization optimization generalized maximum correntropy criterion adaptive filtering (SPOAF-GMCC) algorithm is an online identification method for nonlinear systems that can effectively suppress the interference of impulsive noise. However, as the number of FIR filter weights increases, the computational complexity of the SPOAF-GMCC algorithm in the adaptive process will increase drastically. To improve the computational efficiency of the SPOAF-GMCC algorithm, a nonlinear frequency domain SPOAF-GMCC (FDSPOAF-GMCC) algorithm is proposed in this paper. The range of learning rates when the proposed FDSPOAF-GMCC algorithm converges is theoretically analyzed. In addition, to further improve the convergence speed and reduce the error in the adaptive process, a frequency domain spline prioritization optimization generalized maximum correntropy criterion evolved momentum adaptive filtering (FDSPOAF-GMCC-LMON) algorithm by introducing the evolved momentum (LMON) algorithm is proposed in this paper. Numerical simulations have confirmed that the proposed FDSPOAF-GMCC-LMON algorithm has better robustness to impulsive noise and higher computational efficiency than the frequency domain spline adaptive filtering algorithm. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF