Title | ||
---|---|---|
Identification of fifth-order block-structured nonlinear channels using i.i.d. input signals |
Abstract | ||
---|---|---|
This paper is concerned with the problem of nonlinear Wiener chan- nel identification using input-output crossmoments. The static non- linearity is assumed to be represented by a fifth-degree polynomial. For an i.i.d. input signal, we first derive closed-form expressions for estimating the second-order kernel of the associated fifth-order Volterra model. The parameters of the linear part of the fifth-order Wiener channel are then estimated using an eigenvalue decomposi- tion of the associated second-order Volterra kernel, while the non- linear subsystem is estimated in the least square sense from the re- constructed output of the linear subsystem. The proposed identifi- cation method is illustrated by means of simulation results. |
Year | Venue | Keywords |
---|---|---|
2008 | EUSIPCO | wiener filters,eigenvalues and eigenfunctions,least squares approximations,nonlinear filters,associated fifth-order volterra model,closed-form expressions,eigenvalue decomposition,fifth-degree polynomial,fifth-order wiener channel,fifth-order block-structured nonlinear channels,i.i.d. input signals,input-output crossmoments,least square sense,nonlinear wiener channel identification,nonlinear subsystem,second-order volterra kernel,static nonlinearity |
Field | DocType | ISSN |
Kernel (linear algebra),Least squares,Applied mathematics,Discrete mathematics,Nonlinear system,Expression (mathematics),Polynomial,Communication channel,Independent and identically distributed random variables,Eigendecomposition of a matrix,Mathematics | Conference | 2219-5491 |
Citations | PageRank | References |
1 | 0.35 | 7 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Alain Y. Kibangou | 1 | 95 | 12.01 |
GéRard Favier | 2 | 514 | 46.41 |