Collaborative method based on the acoustical interaction effects on active noise control systems over distributed networks
Autor: | Miguel Ferrer, Alberto Gonzalez, Christian Antoñanzas, Maria de Diego |
---|---|
Rok vydání: | 2017 |
Předmět: |
Computer science
Distributed computing 020206 networking & telecommunications 02 engineering and technology 01 natural sciences Least mean squares filter Frequency domain 0103 physical sciences Path (graph theory) 0202 electrical engineering electronic engineering information engineering Focus (optics) 010301 acoustics Collaborative method Eigenvalues and eigenvectors Active noise control |
Zdroj: | ICASSP |
DOI: | 10.1109/icassp.2017.7952227 |
Popis: | This paper focus on the implementation of an active noise control system over a network of distributed nodes acoustically coupled. We have considered the distributed Multiple Error filtered-x Least Mean Square (DMEFxLMS) algorithm that allows collaboration between nodes following an incremental strategy. To reduce the computational network requirements, an alternative strategy which brings together only the acoustically coupled nodes has been used. However, a rule to define the subsets of nodes that are required to collaborate is needed. A collaborative condition based on the analysis in the frequency domain of the eigenvalues of the acoustic path matrix is proposed. Results demonstrate the ability of the introduced method to define the subsets of nodes that collaborate. |
Databáze: | OpenAIRE |
Externí odkaz: |