.png){kind=icon}
Return
Robust Distributed Adaptation Under Multikernel Correntropy
.png){kind=icon}
PRE
AI
DOI：10.1109/TCSII.2022.3197620.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Most distributed estimation algorithms are hampered by a priori noise knowledge, such as Gaussian distribution. However, there are various noises in many engineering applications, and the distribution of noise is rarely available in advance. Traditional estimators, inevitably, encounter performance degradation simply because their noise assumptions do not hold in practice. To address this problem, we propose a robust diffusion semi-parametric adaptive algorithm for distributed estimation where the adaptive cost function is based on multi-kernel correntropy. Moreover, an online strategy is designed to update variable parameters that constitute the cost function. Simulations in Gaussian and several non-Gaussian noises validate the efficiency and superiority of the proposed algorithm.
Keywords:
Signal processing algorithms
Kernel
Cost function
Estimation
Machine learning algorithms
Uncertainty
Probability density function
Distributed estimation
robustness
diffusion
multi-kernel correntropy
adaptive cost function
non-Gaussian
Journal
I
IF:
4.9
Papers:
8.8K
Citations:
2.5W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveShareSaveShareSaveShareSave
View more