Nonlinear parametric models from volterra kernels measurements

Measuring Volterra Kernels

Volterra series have been in the engineering literature for some time now, and yet there have been few attempts to measure Volterra kernels. This paper discusses techniques for measuring the Volterra kernels of weakly nonlinear systems. We introduce a new quick method for measuring the second Volterra kernel which is analogous to pseudo-noise testing of a linear device. To illustrate the discus...

Estimating Nonlinearity Using Volterra Kernels in Feedback with Linear Models

Aeroelastic dynamics must be accurately known to ensure safe and efficient flight testing. Unfortunately, most models of aircraft systems typically describe only the linear dynamics. These models are inadequate for predicting behaviors, such as limit cycle oscillations, resulting from nonlinearities. This paper presents an approach to augment a linear model by identifying associated nonlinear o...

Nonlinear Parametric Hidden Markov Models

In previous work [4], we extended the hidden Markov model (HMM) framework to incorporate a global parametric variation in the output probabilities of the states of the HMM. Development of the parametric HMM was motivated by the task of simultaneoiusly recognizing and interpreting gestures that exhibit meaningful variation. With standard HMMs, such global variation confounds the recognition proc...

Identification of Nonlinear Models by Input-output Measurements: the Volterra Series Approach

We propose the identification method based on input-output measurements. Our approach exploits the Volterra functional series expansion for empirical “black-box” characterization of non-linear dynamic systems with memory. Procedure of transfer functions identification involves numerical solution of the special Volterra integral equations derived by the parametric family of input sequences with ...

Volterra kernels and effective connectivity