📄 eigen_filtering.m
字号:
function [s,w,g] = eigen_filtering(y,p,mu);% Given a 1D noisy sequence y, the order p of% the ARMA(p,p) model and the regularization parameter mu% this function computes the clean signal s, an estimate of % the noise sequence, and the predicor error fiter g. N = length(y); Y = convmtx(y,p+1); % data convolution matrix R = Y'*Y/N; % data correlation matrix [g,Pw] = eigs(R,1,'SM'); % compute 1 eigenvalue (the SMallest) g0 = g(1) g = g/g0; e = Y*g; % non-white sequence e G = convmtx(g,N); % correlation matrix of the filter D = eye(N)*mu; % regularization term w = inv(G'*G+D)*G'*e; % estimate of the noise s = y-w; % estimate of the clean signal return
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -