TY - GEN
T1 - Distributed nonlinear regression using in-network processing with multiple Gaussian kernels
AU - Shin, Ban Sok
AU - Paul, Henning
AU - Yukawa, Masahiro
AU - Dekorsy, Armin
N1 - Funding Information:
ACKNOWLEDGMENT The work leading to this publication was partially funded by the German Research Foundation (DFG) under grant Pa2507/1. M. Yukawa is thankful to JSPS Grants-in-Aid (15K06081, 15K13986, 15H02757).
Publisher Copyright:
© 2017 IEEE.
PY - 2017/12/19
Y1 - 2017/12/19
N2 - In this paper, we propose the use of multiple Gaussian kernels for distributed nonlinear regression or system identification tasks by a network of nodes. By employing multiple kernels in the estimation process we increase the degree of freedom and thus, the ability to reconstruct nonlinear functions. For this, we extend the so-called KDiCE algorithm, which allows a distributed regression of nonlinear functions but uses a single kernel only, to multiple kernels. We corroborate our proposed scheme by numerical evaluations for the reconstruction of nonlinear functions both static and time-varying. We achieve performance gains for both cases, in particular for the tracking of a time-varying nonlinear function.
AB - In this paper, we propose the use of multiple Gaussian kernels for distributed nonlinear regression or system identification tasks by a network of nodes. By employing multiple kernels in the estimation process we increase the degree of freedom and thus, the ability to reconstruct nonlinear functions. For this, we extend the so-called KDiCE algorithm, which allows a distributed regression of nonlinear functions but uses a single kernel only, to multiple kernels. We corroborate our proposed scheme by numerical evaluations for the reconstruction of nonlinear functions both static and time-varying. We achieve performance gains for both cases, in particular for the tracking of a time-varying nonlinear function.
KW - Distributed regression
KW - In-network processing
KW - Kernel least-squares
KW - Multiple kernels
UR - http://www.scopus.com/inward/record.url?scp=85044203147&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044203147&partnerID=8YFLogxK
U2 - 10.1109/SPAWC.2017.8227645
DO - 10.1109/SPAWC.2017.8227645
M3 - Conference contribution
AN - SCOPUS:85044203147
T3 - IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC
SP - 1
EP - 5
BT - 18th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 18th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2017
Y2 - 3 July 2017 through 6 July 2017
ER -