Back to results
US 2013/0173678 A1Published application
Multi-stage adaptive filter
AssigneeIntel Corporation
InventorYANG-SEOK CHOI
Life of this patent
Estimating the term…
Abstract
A multi-stage adaptive filter is disclosed, which exhibits a smaller mean square error than in prior art adaptive filters. The adaptive filter selectively manipulates the weights, in multiple stages, so as to achieve a global minimum of the error function, such that the filtered signal has as small an error as possible.
Claims20
A method, comprising:
- running an adaptive filter on an incoming digital signal being received into a system, the adaptive filter to generate an output signal, the adaptive filter to receive a weight vector, w, the weight vector comprising a plurality of weights, w=[wi1,wi2, . . . , wiN]T, received into the adaptive filter, wherein the adaptive filter is executed for a predetermined number of iterations, the plurality of weights having original values, |wi1|, |wi2|, . . . , |wiN|, following the iterations;
- ordering the original values, |wi1|, |wi2|, . . . , |wiN|, of the plurality of weights, w=[wi1,wi2, . . . , wiN]T, according to a formula, |wik|≧|wik+1|, for index, ik, to produce a new weight vector;
- setting each of the plurality of weights of the new weight vector to zero;
- and
- for k=1 to N:
- updating the k weights, wi1, wi2 . . . wik, of the new weight vector;
- and
- repeatedly running the adaptive filter using the new weight vector until the k weights, wi1, wi2 . . . wik, converge.
5 more elements
A receiver, comprising:
- an adaptive filter to receive an incoming digital signal and to generate a new digital signal;
- and
- a software program executed by a processor, the software program to control the execution of the adaptive filter, the software program to further feed a plurality of weights into the adaptive filter, wherein the plurality of weights is denoted by, w=[wi1,wi2, . . . , wiN]T;
- wherein the software program:
- executes the adaptive filter for a predetermined number of iterations, such that the plurality of weights, following the iterations, have values, |wi1|, |wi2|, . . . , |wiN|;
- orders the values, |wi1|, |wi2|, . . . , |wik|, of the plurality of weights, wi1,wi2, . . . , wiN, according to a formula, for index, |wik|≧|wik+1|, for index, ik;
- generates a new weight vector by setting each of the plurality of weights of the weight vector to zero;
- and
- for k=1 to N:
- updates the k weights, wi1, wi2 . . . wik, of the new weight vector;
- and
- repeatedly runs the adaptive filter using the new weight vector until the k weights, wi1, wi2 . . . wik, converges.
9 more elements
An article comprising a medium storing a software program which, when executed by a processor, performs the following operations:
- generate a plurality of weights to be received by an adaptive filter, the plurality of weights being denoted by, w=[wi1,wi2, . . . , wiN]T;
- execute the adaptive filter on an incoming digital signal for a predetermined number of iterations, wherein the plurality of weights have values, |wi1|, |wi2|, . . . , |wiN|, following the iterations;
- order the values, |wi1|, |wi2|, . . . , |wiN|, of the plurality of weights, w1,w2, . . . , wN according to a formula, |wik|≧|wik+1|, for index, ik;
- generate a new weight vector by setting each of the plurality of weights of the weight vector to zero;
- for k=1 to N:
- update the k weights, wi1, wi2 . . . wik, of the new weight vector;
- and
- repeatedly run the adaptive filter using the new weight vector until the k weights, wi1, wi2 . . . wik, converges;
- wherein an output signal is generated that has fewer errors than the incoming digital signal.
6 more elements
Dependencies are inferred from the claim text.