TY - JOUR
T1 - Robust and low complexity distributed kernel least squares learning in sensor networks
AU - Pérez-Cruz, Fernando
AU - Kulkarni, Sanjeev R.
N1 - Funding Information:
Manuscript received November 30, 2009; revised January 07, 2010. First published January 22, 2010; current version published February 17, 2010. This work was supported in part by the U.S. Office of Naval Research under Grant N00014-07-1-0555, the U.S. Army Research Office under Grant W911NF-07-1-0185, the Spanish government (TEC2006-13514-C02-01/TCM, TEC2009-14504-C02-01, and CSD2008-00010). The work of F. Pérez-Cruz was supported by Marie Curie Fellowship 040883-AI-COM. This work was presented in part at ISIT 2009 [1]. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Hsiao-Chun Wu.
PY - 2010
Y1 - 2010
N2 - We present a novel mechanism for consensus building in sensor networks. The proposed algorithm has three main properties that make it suitable for sensor network learning. First, the proposed algorithm is based on robust nonparametric statistics and thereby needs little prior knowledge about the network and the function that needs to be estimated. Second, the algorithm uses only local information about the network and it communicates only with nearby sensors. Third, the algorithm is completely asynchronous and robust. It does not need to coordinate the sensors to estimate the underlying function and it is not affected if other sensors in the network stop working. Therefore, the proposed algorithm is an ideal candidate for sensor networks deployed in remote and inaccessible areas, which might need to change their objective once they have been set up.
AB - We present a novel mechanism for consensus building in sensor networks. The proposed algorithm has three main properties that make it suitable for sensor network learning. First, the proposed algorithm is based on robust nonparametric statistics and thereby needs little prior knowledge about the network and the function that needs to be estimated. Second, the algorithm uses only local information about the network and it communicates only with nearby sensors. Third, the algorithm is completely asynchronous and robust. It does not need to coordinate the sensors to estimate the underlying function and it is not affected if other sensors in the network stop working. Therefore, the proposed algorithm is an ideal candidate for sensor networks deployed in remote and inaccessible areas, which might need to change their objective once they have been set up.
KW - Consensus
KW - Distributed learning
KW - Kernel methods
KW - Sensor networks
UR - http://www.scopus.com/inward/record.url?scp=77249098741&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77249098741&partnerID=8YFLogxK
U2 - 10.1109/LSP.2010.2040926
DO - 10.1109/LSP.2010.2040926
M3 - Article
AN - SCOPUS:77249098741
VL - 17
SP - 355
EP - 358
JO - IEEE Signal Processing Letters
JF - IEEE Signal Processing Letters
SN - 1070-9908
IS - 4
M1 - 5395679
ER -