The projection filter is an approximate nonlinear filter based on orthogonal projection in the tangent space of a manifold of densities. The Riemannian metric used is the Fisher information metric. In this paper it is shown that if one uses Gaussian densities the projection filter equals a McShane-Fisk-Stratonovich version of the socalled Assumed Density Filter.

The nonlinear speech signal decomposition based on Volterra-Wiener functional series is described. The nonlinear filter bank structure is proposed for phonemes recognition solving.

Report [1] compares the Extended Kalman Filter [2, 3, 4], the Iterated Extended Kalman Filter, IEKF, [2, 3, 4] and the Linear Regression Kalman Filter [5] (e.g. the Unscented Kalman Filter, UKF, [6, 7, 8]) on (i) consistency and (ii) information content of their results (estimates and covariance matrices). The nonlinear filter proposed by Bellaire et al. in [9] is not discussed in report [1]. T...

four explicit finite difference schemes, including lax-friedrichs, nessyahu-tadmor, lax-wendroff and lax-wendroff with a nonlinear filter are applied to solve water hammer equations. the schemes solve the equations in a reservoir-pipe-valve with an instantaneous and gradual closure of the valve boundary. the computational results are compared with those of the method of characteristics (moc), a...

Orbit determination in application to the estimation of impact probability has the goal of determining the evolution of the state probability density function (pdf) and determining a measure of the probability of collision. Nonlinear gravitational interaction and non-conservative forces can make the pdf far from Gaussian. This work implements three nonlinear sequential estimators: the Extended ...

EGG, which depicts the degree of contact between vocal folds, is a measurable proxy for glottal pressure wave, and thus contains information of pitch and GCI. In this project, I proposed a TDRNN, or a nonlinear ARMA filter, that turns input speech waveforms to approximation of EGG, which can be further applied to pitch tracking or GCI location. Two baseline systems, a linear MA filter and a non...

In this work we predict the nonlinear behaviour of an eight-pole quasi-elliptic bandpass high temperature superconducting (HTS) filter with an equivalent circuit extracted from intermodulation measurements performed at the centre of the filter passband. We present measurements that show that the equivalent circuit is able to predict the intermodulation products produced by the filter when drive...

The Bayesian approach provides a rather powerful framework for handling nonlinear, as well as linear, estimation problems. We can in fact pose a general solution to the nonlinear estimation problem. However, in the general case there does not exist any closed-form solution and we are forced to use approximate techniques. In this thesis we will study one such technique, the sequential Monte Carl...

A generalized dynamical robust nonlinear filtering framework is established for a class of Lipschitz differential algebraic systems, in which the nonlinearities appear both in the state and measured output equations. The system is assumed to be affected by norm-bounded disturbance and to have both norm-bounded uncertainties in the realization matrices as well as nonlinear model uncertainties. W...

Distributed linear estimation theory has received increased attention in recent years due to several promising industrial applications. Distributed nonlinear estimation, however is still a relatively unexplored field despite the need in numerous practical situations for techniques that can handle nonlinearities. This paper presents a unified way of describing distributed implementations of thre...