.png){kind=icon}
Return
Diffusion Mixture Minimum Total Error Entropy Adaptive Filtering Algorithm and Its Performance Analysis
.png){kind=icon}
PRE
AI
DOI：10.1109/TSIPN.2023.3288304.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The distributed adaptive filtering algorithms applied in the error-in-variables (EIV) model consider that both input and output signals suffer from noise disturbance for practical applications. However, the traditional distributed adaptive filtering algorithms applied in the EIV model cannot combat non-Gaussian noises well. To address this issue, this article develops a mixture minimum total error entropy (MMTE) method and its diffusion mixture minimum total error entropy (DMMTE) algorithm. The proposed DMMTE has the significant performance improvement in non-Gaussian noises, and is more general than the diffusion minimum total error entropy (DMTE) algorithm. To select mixture coefficient effectively, a variable mixture coefficient method is further proposed to generate a variable DMMTE (VDMMTE) algorithm. Moreover, the local mean stability analysis and steady-state performance analysis of DMMTE are presented. Simulations verify the correctness of the theoretical analysis and illustrate the performance superiority of the proposed DMMTE.
Keywords:
Distributed estimation
mixture minimum error entropy
adaptive filter
errors-in-variables model
total least squares
Journal
IF:
4.9
Papers:
736
Citations:
1.9K
Organization
Cited Papers
.png){kind=icon}
ShareSave
ShareSaveShareSaveShareSave
View more