Topic
Kernel adaptive filter
About: Kernel adaptive filter is a research topic. Over the lifetime, 8771 publications have been published within this topic receiving 142711 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: This article presents a novel adaptive harmonic IIR notch filter with a single adaptive coefficient to efficiently perform frequency estimation and tracking in a harmonic frequency environment and devise a simple scheme to select the initial filter coefficient to insure algorithm convergence to its global minimum error.
Abstract: In many applications, a sinusoidal signal may be subjected to nonlinear effects in which possible harmonic frequency components are generated. In such an environment, we may want to estimate (track) the signal's fundamental frequency as well as any harmonic frequencies. Using a secondorder notch filter to estimate fundamental and harmonic frequencies is insufficient since it only accommodates one frequency component. On the other hand, applying a higher-order infinite impulse response (IIR) notch filter may not be efficient due to adopting multiple adaptive filter coefficients. In this article, we present a novel adaptive harmonic IIR notch filter with a single adaptive coefficient to efficiently perform frequency estimation and tracking in a harmonic frequency environment. Furthermore, we devise a simple scheme to select the initial filter coefficient to insure algorithm convergence to its global minimum error.
41 citations
TL;DR: The proposed wavelet-based method was applied to limbic P300 potentials and variance of single trial MTL-P300s decreased, without restricting the corresponding mean, and can be regarded as an alternative for single-trial ERP analysis.
Abstract: We present a new wavelet-based method for single trial analysis of transient and time variant event-related potentials (ERPs). Expecting more accurate filter settings than achieved by other techniques (low-pass filter, a posteriori Wiener filter, time invariant wavelet filter), ERPs were initially balanced in time. By simulation, better filter performance could be established for test signals contaminated with either white noise or isospectral noise. To provide an example of real application, the method was applied to limbic P300 potentials (MTL-P300). As a result, variance of single trial MTL-P300s decreased, without restricting the corresponding mean. The proposed method can be regarded as an alternative for single-trial ERP analysis.
41 citations
24 Nov 2003
TL;DR: Kernel particle filter is presented as a variation of particle filter with improved sampling efficiency and performance in visual tracking by invoking kernel-based representation of densities and introducing mean shift as an iterative mode-seeking procedure.
Abstract: Particle filter has recently received attention in computer vision applications due to attributes such as its ability to carry multiple hypotheses and its relaxation of the linearity assumption. Its shortcoming is increase in complexity with state dimension. We present kernel particle filter as a variation of particle filter with improved sampling efficiency and performance in visual tracking. Unlike existing methods that use stochastic or deterministic optimization procedures to find the modes in a likelihood function, we redistribute particles by invoking kernel-based representation of densities and introducing mean shift as an iterative mode-seeking procedure, in which particles move towards dominant modes while still maintaining as fair samples from the posterior. Experiments on face and limb tracking show that the algorithm is superior to conventional particle filter in handling weak dynamic models and occlusions with 60% fewer particles in 3-9 dimensional spaces.
41 citations
TL;DR: The L1 adaptive controller ensures that the nonlinear, affine-in-control, uncertain system follows its ideal nonlinear model during the transient and steady-state, if the adaptation gain is selected sufficiently large and the bandwidth of the low-pass filter in the L2 adaptive control architecture is adjusted appropriately.
41 citations
Patent•
06 May 1994TL;DR: In this paper, the adaptive control unit adjusts the characteristic of the filter on the basis of a decode error (residual) at the decoder 15 and an input to the filter 14.
Abstract: The characteristic of a regenerative signal from a magnetic head 11 is compensated by a filter 14 serving as an equalizer. The regenerative signal thus compensated is then decoded at a decoder 15. An adaptive control unit 17 adjusts (modifies) the characteristic of the filter 14 on the basis of a decode error (residual) at the decoder 15 and an input to the filter 14. A servo control unit 18 sends a servo lock signal to the adaptive control unit 17 when a servo control operation at the time of reproduction is stabilized to start an automatic adjustment operation of the filter characteristic. Thus, it is possible to prevent in advance bad influence or effect on a compensating operation of the equalizer resulting from the fact that an adaptive adjustment operation of the filter characteristic might be carried out at the time when the servo control operation is unstable like at the time of building up of a reproducing operation.
41 citations