%0 Journal Article %1 KimJeongEtAl09puc %A Kim, Youngsoo %A Jeong, Sangbae %A Kim, Daeyoung %A López, Tomás Sánchez %D 2009 %J Personal and Ubiquitous Computing %K v1205 springer paper embedded ai sensor network data processing pattern recognition information retrieval learn %N 7 %P 499-508 %R 10.1007/s00779-009-0225-8 %T An Efficient Scheme of Target Classification and Information Fusion in Wireless Sensor Networks %V 13 %X In this paper, an efficient target classification and fusion scheme for wireless sensor networks (WSNs) is proposed and evaluated. When a classification algorithm for WSN nodes is designed, parametric approaches such as Gaussian mixture model (GMM) should be more preferred to non-parametric ones due to the hard limitation in resources. The GMM algorithm not only shows good performances for target classification in WSNs but it also requires very small resources. Based on the classifier, a decision tree generated by the classification and regression tree algorithm is used to fuse the information from heterogeneous sensors. This node-level classification scheme provides a satisfactory classification rate, 94.10 percent, with little resources. Finally, a confidence-based fusion algorithm improves the overall accuracy by fusing the information among sensor nodes. Our experimental results show that the proposed group-level fusion algorithm improves the accuracy by an average of 4.17 percent accuracy with randomly selected nodes.

@article{KimJeongEtAl09puc, abstract = {In this paper, an efficient target classification and fusion scheme for wireless sensor networks (WSNs) is proposed and evaluated. When a classification algorithm for WSN nodes is designed, parametric approaches such as Gaussian mixture model (GMM) should be more preferred to non-parametric ones due to the hard limitation in resources. The GMM algorithm not only shows good performances for target classification in WSNs but it also requires very small resources. Based on the classifier, a decision tree generated by the classification and regression tree algorithm is used to fuse the information from heterogeneous sensors. This node-level classification scheme provides a satisfactory classification rate, 94.10 percent, with little resources. Finally, a confidence-based fusion algorithm improves the overall accuracy by fusing the information among sensor nodes. Our experimental results show that the proposed group-level fusion algorithm improves the accuracy by an average of 4.17 percent accuracy with randomly selected nodes.}, added-at = {2012-05-30T10:49:12.000+0200}, author = {Kim, Youngsoo and Jeong, Sangbae and Kim, Daeyoung and L\'{o}pez, Tom\'{a}s S\'{a}nchez}, biburl = {https://www.bibsonomy.org/bibtex/2dd3ef288593cd91ce912e60f01ff14ca/flint63}, doi = {10.1007/s00779-009-0225-8}, file = {SpringerLink:2009/KimJeongEtAl09puc.pdf:PDF}, groups = {public}, interhash = {0f20e1e13d50f029d33c200753631acd}, intrahash = {dd3ef288593cd91ce912e60f01ff14ca}, issn = {1617-4909}, journal = {Personal and Ubiquitous Computing}, keywords = {v1205 springer paper embedded ai sensor network data processing pattern recognition information retrieval learn}, number = 7, pages = {499-508}, timestamp = {2018-04-16T11:44:59.000+0200}, title = {An Efficient Scheme of Target Classification and Information Fusion in Wireless Sensor Networks}, username = {flint63}, volume = 13, year = 2009 }