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.

A system identification algorithm using orthogonal functions

Abstract: An adaptive filter (ADF) structure is proposed for applications in which large-order ADFs are required. It is based on modeling the impulse response of the system to be identified as a linear combination of a set of discrete Legendre orthogonal functions. The proposed adaptive filter structure has desirable stability features and a unimodal mean-square error surface as well as a modular structure that permits an easy increase of the filter order without changing the previous stages. Computer simulations in which the proposed structure is used to identify actual acoustic echo path impulse responses show that the Legendre ADF has better convergence performance than the transversal ADF when identifying systems with long impulse response. >

Range Sidelobe Reduction Filter Design for Binary Coded Pulse Compression System

Abstract: Pulse compression is an active research topic in radar systems. A problem in pulse compression is masking of small targets by the range sidelobes of large nearby targets. A new filter structure is introduced here to suppress the sidelobes of radar signals that result from standard matched filtering. The proposed filter is applicable for any type of binary coding signals. Several techniques are used to calculate the filter coefficients such as Wiener filter technique, Lagrange multiplier method, and linear programming (LP) algorithm. Also, a weighting function is utilized to shape the sidelobe energy in an iterative manner that will yield more sidelobe reduction. Comparison of the proposed filters and the matched filter shows that at the expense of insignificant loss in signal-to-noise ratio (LSNR), adequate mainlobe-to-peak-sidelobe ratio (MSR) can be achieved.

Adaptive equalizing apparatus and program therefor

Abstract: In an adaptive equalizing apparatus for MIMO (Multi-Input Multi-Output) turbo reception, an interference component in a received signal is subtracted therefrom using a replica of an interference component in an interference canceling part 31 n , the subtracted output is filtered by a filter 32 n , to cancel the remaining interference component and to perform multi-path combining, and in a degree-of-interference-cancellation estimation part 41 n the degree of interference cancellation β(i) is set such that it is 0 for the iteration number i=1, 0.8+0.05 (i−1) for 5≧i≧2 and 1 for i≧6, and at the beginning of each iteration filter coefficients are calculated using β(i) and a channel estimation value in a filter coefficient calculating part 33 n and the filter coefficient thus calculated are set in the filter 32 n . An average value of soft decision symbol estimation values used in the interference canceling part may be used as β.

Wideband adaptive prediction

Abstract: Superior performance, better tracking, and stability of an adaptive predictor are realized by employing a so-called high order (eighth) prediction filter and by updating the filter coefficients in accordance with an algorithm based on the sign of a reconstructed signal sampel (sq(k)) multiplied by a quantized version (eq(k)) of a differential error signal normalized by an expected magnitude of the reconstructed signal sample.

Simultaneous optimisation of the low-pass filter and state-feedback controller in a robust repetitive-control system

Abstract: A low-pass filter is inserted in a repetitive controller to guarantee the stability of the modified repetitive-control system. The control precision strongly depends on the parameter of the filter. This study presents a method of simultaneously optimising the parameters of the low-pass filter and state feedback of a modified repetitive-control system in which the plant contains a class of uncertainties. First, the relationship between the control precision of a repetitive-control system and a low-pass filter is explained. Next, a linear matrix inequality (LMI)-based robust-stability condition is derived for fixed state-feedback gains. This condition is transformed into a generalised eigenvalue problem and is used to calculate the maximum cut-off angular frequency of the low-pass filter. Then, another LMI-based robust-stability condition is derived for a fixed low-pass filter, and is employed to find H∞ static-state-feedback gains. Moreover, an iterative algorithm that combines these two robust-stability conditions is designed that yields the largest bandwidth while guaranteeing closed-loop robust stability. The conservativeness of the result produced by the algorithm is the same as that of the less conservative of the two robust-stability conditions. Finally, two numerical examples demonstrate the validity of the method.