Wavelet based detection of changes in the composition of RLC networks

Dielectrics 2013

Paiva, H.M.^1 ; Duarte, M.A.Q.^2 ; Galvao, R.K.H.^3 ; Hadjiloucas, S.^4

Mectron-Odebrecht Organization São José Dos Campos, SP, 12227-000, Brazil^1

Department of Mathematics, Universidade Estadual de Mato Grosso Do sul (UEMS), 79540-000, Cassilandia, MS, Brazil^2

Divisão de Engenharia Eletrônica, Instituto Tecnológico de Aeronáutica, São José dos Campos, SP, 12228-900, Brazil^3

School of Systems Engineering, University of Reading, RG6 6AY, United Kingdom^4

关键词: 3-dimensional;    Amorphous dielectrics;    Chemometric classification;    Compositional analysis;    Drude response;    Input-output transfer functions;    Wavelet based detection;    Wavelet filter banks;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/472/1/012011/pdf
DOI  :  10.1088/1742-6596/472/1/012011

来源: IOP

【 摘 要 】

The current work discusses the compositional analysis of spectra that may be related to amorphous materials that lack discernible Lorentzian, Debye or Drude responses. We propose to model such response using a 3-dimensional random RLC network using a descriptor formulation which is converted into an input-output transfer function representation. A wavelet identification study of these networks is performed to infer the composition of the networks. It was concluded that wavelet filter banks enable a parsimonious representation of the dynamics in excited randomly connected RLC networks. Furthermore, chemometric classification using the proposed technique enables the discrimination of dielectric samples with different composition. The methodology is promising for the classification of amorphous dielectrics.