.png){kind=icon}
.png){kind=icon}
Return
Minimum Error Entropy Based Diffusion Normalized Euclidean Direction Search Algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jfranklin.2025.107953.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Based on the Lagrangian methodology and the minimum error entropy (MEE) criterion, a novel distributed MEE normalized Euclidean direction search (MEE-DNEDS) algorithm is proposed. Benefiting from the MEE and the EDS algorithm, the MEE-DNEDS algorithm can effectively combat impulsive noise and improve tracking performance. To reduce the communication burden, a communication reducing MEE-DNEDS (CR-MEE-DNEDS) algorithm is further proposed, which employs the smart selective method to enhance robustness and achieve reduced communication cost. In addition, the mean convergence performance of the CR-MEE-DNEDS algorithm is analyzed. Simulation results demonstrate the improved performance of the proposed algorithms for distributed in-network system identification and stereophonic acoustic echo cancellation as compared to the state-of-the-art algorithms.
Keywords:
Lagrangian methodology
minimum error entropy
distributed algorithm
impulsive noise
communication reducing
Journal
J
IF:
4.2
Papers:
822
Citations:
0