.png){kind=icon}
Return
Proportionate Maximum Versoria Criterion-Based Adaptive Algorithm for Sparse System Identification
.png){kind=icon}
PRE
AI
DOI：10.1109/TCSII.2021.3123055.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Proportionate Maximum Versoria Criterion (P-MVC) based adaptive algorithms for unknown sparse system identification problem are proposed in this brief. The conventional proportionate type algorithms used for sparse system identification can work well only under Gaussian assumption due to the dependency on the least mean square error. However, in many real cases, the algorithms have to be also robust in impulsive noise environments. The Maximum Versoria Criteria based adaptive algorithms were found to have good robustness against impulsive noise while the proportionate term in the adaptive algorithm exploits the sparse nature to improve the convergence speed. Hence, to simultaneously have robustness under impulsive environment and improved convergence speed, the P-MVC algorithm and an improved tracking P-MVC version are proposed. The performance analysis indicates that the Excess Mean Square Error (EMSE) is the same as that of MVC adaptive algorithm. Furthermore, simulations in the context of sparse system identification scenario reveal that the proposed algorithms have both robustness and improved performance in impulsive noise environment.
Keywords:
Adaptive filters
System identification
Steady-state
Convergence
Mean square error methods
Finite impulse response filters
Filtering algorithms
Adaptive filters
system identification
steady-state
convergence
mean square error methods
finite impulse response filters
Journal
I
IF:
4.9
Papers:
8.8K
Citations:
2.5W
Organization
Cited Papers
.png){kind=icon}
ShareSave
ShareSave
Steady State Mean Square Analysis of Standard Maximum Versoria Criterion Based Adaptive Algorithm
PREAI
ShareSaveShareSave
View more