Weighted Tensor Decomposition for Approximate Decoupling of Multivariate Polynomials

In the field of system identification, one special type of nonlinear models are the so-called block-oriented models, consisting of linear time-invariant blocks and nonlinear static blocks. More specifically, this presentation will focus its attention on the parallel Wiener-Hammerstein models. When identifying these types of systems, a multiple-input-multiple-output polynomial should be decoupled, that was obtained from noisy measurements. In our work, an earlier developed decoupling algorithm with good results in the noiseless case is generalized to the noisy case.