.png){kind=icon}
.png){kind=icon}
Return
A Novel Combinatoric Correntropy Algorithm: Properties and its Performance Analysis
.png){kind=icon}
PRE
AI
DOI：10.1109/TCSII.2022.3200417.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In order to improve the flexibility of the correntropy that is usually in the form of a single kernel, this brief proposes a novel combinatoric maximum correntropy criterion (C-MCC), where the combinatoric correntropy (C-C) consists of two different single-type kernels. As a new form of correntropy, the C-C has a more generalized form than the traditional single-kernel correntropy, and the corresponding properties and proofs of the C-C are given in this brief. It is noticed that the combined kernel can become a single kernel by changing the combined parameters. Moreover, the adaptive filtering algorithm based on C-MCC is derived and the theoretical performance analysis of the C-MCC algorithm is also provided. Finally, the experimental simulations show that the proposed algorithm has better convergence speed and steady-state performance than MCC-type and other algorithms, and the proposed method of the theoretical analysis of C-MCC agrees well with simulation results.
Keywords:
Adaptive filtering
combinatoric maximum correntropy
combined parameter
Journal
I
IF:
4.9
Papers:
8.8K
Citations:
2.5W