Sökning: "Nonlinear filters"

Visar resultat 1 - 5 av 45 avhandlingar innehållade orden Nonlinear filters.

  1. 1. System Modeling Using Basis Functions and Application to Echo Cancelation

    Författare :Lester S.H. Ngia; Chalmers University of Technology; []
    Nyckelord :Laguerre filters; Gauss-Newton algorithms; Levenberg-Marquard algorithms; room acoustics; separable nonlinear least-squares; adaptive filtering; off-line estimation; echo cancelation; on-line estimation; system modeling; neural net filters; Kautz filters; loudspeaker systems; steepest-descent algorithms; Hammerstein models;

    Sammanfattning : This thesis concerns modeling of linear and nonlinear dynamic systems, where the applied models can be described by basis function expansions. In the first part of the thesis, the investigated model (or filter) structures and parameter estimation algorithms are described. LÄS MER

  2. 2. Identification of Stochastic Nonlinear Dynamical Models Using Estimating Functions

    Författare :Mohamed Abdalmoaty; Håkan Hjalmarsson; Adrian Wills; KTH; []
    Nyckelord :Prediction Error Method; Maximum Likelihood; Data-driven; Learning; Stochastic; Nonlinear; Dynamical Models; Non-stationary Linear Predictors; Intractable Likelihood; Latent Variable Models; Estimation; Process Disturbance; Electrical Engineering; Elektro- och systemteknik;

    Sammanfattning : Data-driven modeling of stochastic nonlinear systems is recognized as a very challenging problem, even when reduced to a parameter estimation problem. A main difficulty is the intractability of the likelihood function, which renders favored estimation methods, such as the maximum likelihood method, analytically intractable. LÄS MER

  3. 3. Generation and Applications of Picosecond Pulses in High-Capacity Fiber Optic Systems

    Författare :Jie Li; Chalmers University of Technology; []
    Nyckelord :nonlinear distortion; optical pulse shaping; dispersion-management; eye-diagram; cross-phase modulation; optical sampling; optical fiber applications; multiplexing; parametric amplification; demultiplexing; optical pulses; fiber optical communication; nonlinear systems; time division multiplexing; four-wave mixing; polarization-mode dispersion; optical signal processing; fiber optical nonlinearities; fiber ring laser; soliton robustness; mode-locking; optical filters; networks;

    Sammanfattning : This thesis deals with short optical pulse generation from actively mode-locked Erbium-doped fiber ring lasers (ML-EFRL) and their applications in all-optical signal processing (optical sampling and switching) and transmission experiments for high-capacity fiber optic systems. In future ultra-high bit-rate optical time division multiplexed (OTDM) systems, one of the key requirements is to develop reliable optical pulse sources that can generate high quality optical pulses. LÄS MER

  4. 4. Kalman Filters for Nonlinear Systems and Heavy-Tailed Noise

    Författare :Michael Roth; Fredrik Gustafsson; Simo Särkkä; Linköpings universitet; []
    Nyckelord :;

    Sammanfattning : This thesis is on filtering in state space models. First, we examine approximate Kalman filters for nonlinear systems, where the optimal Bayesian filtering recursions cannot be solved exactly. These algorithms rely on the computation of certain expected values. LÄS MER

  5. 5. Particle filters and Markov chains for learning of dynamical systems

    Författare :Fredrik Lindsten; Thomas B. Schön; Lennart Ljung; Fredrik Gustafsson; Arnaud Doucet; Linköpings universitet; []
    Nyckelord :ENGINEERING AND TECHNOLOGY; TEKNIK OCH TEKNOLOGIER; NATURAL SCIENCES; NATURVETENSKAP; NATURVETENSKAP; TEKNIK OCH TEKNOLOGIER; NATURAL SCIENCES; ENGINEERING AND TECHNOLOGY; Bayesian learning; System identification; Sequential Monte Carlo; Markov chain Monte Carlo; Particle MCMC; Particle filters; Particle smoothers;

    Sammanfattning : Sequential Monte Carlo (SMC) and Markov chain Monte Carlo (MCMC) methods provide computational tools for systematic inference and learning in complex dynamical systems, such as nonlinear and non-Gaussian state-space models. This thesis builds upon several methodological advances within these classes of Monte Carlo methods. LÄS MER