Adaptive filter
AssigneeQinetiq Limited
InventorsIan David Skidmore, Ian Keith Proudler
Life of this patent
Estimating the term…
Abstract
An adaptive filter is implemented by a computer (10) processing an input signal using a recursive least squares lattice (RLSL) algorithm (12) to obtain forward and backward least squares prediction residuals. A prediction residual is the difference between a data element in a sequence of elements and a prediction of that element from other sequence elements. Forward and backward residuals are converted at (14) to interpolation residuals which are unnormalized Kalman gain vector coefficients. Interpolation residuals are normalized to produce the Kalman gain vector at (16). The Kalman gain vector is combined at (18) with input and reference signals x(t) and y(t), which provides updates for the filter coefficients or weights to reflect these signals as required to provide adaptive filtering.
Claims29
An adaptive filter implemented by means of filter weights, the filter including means for providing a sequence of signal samples for application to the filter and means for updating the filter weights by means of a gain vector derived from interpolation residuals of the sequence of signal samples applied to the filter.
A method for adaptive filtering with filter weights comprising the steps of:
- a) deriving interpolation residuals from a sequence of signal samples provided as data for filtering, and
- b) updating the filter weights with a gain vector derived from the interpolation residuals.
A method for adaptive filtering including:
- a) processing an input sequence of signal samples to derive prediction residuals;
- b) converting prediction residuals to interpolation residuals;
- c) deriving elements of a gain vector from the interpolation residuals;
- and
- d) combining the gain vector with input and reference signals and update the filter coefficients or weights as required to provide adaptive filtering.
A method according 4 claim wherein the prediction residuals are least squares prediction residuals.
A computer software product comprising a computer readable medium containing computer readable instructions for controlling operation of computer apparatus to implement an adaptive filter, wherein the computer readable instructions provide a means for controlling the computer apparatus to input a sequence of signal samples as data for filtering, and the computer readable instructions also provide a means for controlling the computer apparatus to generate updated filter weights by means of a gain vector derived from interpolation residuals of sequence of signal samples.
A computer software product comprising readable medium containing computer readable instructions for controlling operation of computer apparatus to implement an adaptive filter wherein the computer readable instructions provide a means for controlling the computer apparatus to:
- m) process an input sequence of signal samples to derive prediction residuals;
- n) convert prediction residuals to interpolation residuals corresponding to elements of a gain vector;
- and
- o) derive elements of a gain vector from the interpolation residuals;
- and
- p) combine the gain vector with input and reference signals and update the filter coefficients or weights as required to provide adaptive filtering.
Dependencies are inferred from the claim text.