wienerfiltering.m

AR模型的源程序。

% File: WienerFiltering.m
% -----------------------
% This file integrates the sequence generation module and correlation
% calculation module.

function WienerFiltering(L, N)
% L: Sample length
% N: Filter length

WienerFilter_SeqGen(L, N); % Sequence Generation Module
                           % for sample length L = 500,
                           % filter length N = 10.
WienerFilter_Core(); % Correlation Calculation Module

load WienerFilter_SeqGen.mat;
load WienerFilter_Core.mat;

% FIR Wiener Filtering
output_vector_fir = conv(signalvector_x, h_fir); % FIR convolution L+N-1 samples
output_fir = zeros(L, 1);
for n = 1: L
    output_fir(n) = output_vector_fir(n); % L samples cutoff
end

% Ideal Wiener Filtering
%signalvector_x_cat = vertcat(signalvector_x, zeros(N - 1, 1));
s_ide_0 = 0; % for s_ide(1) = 0 + 0.238 x(1)
%output_vector_ide(1) = 0.724 * s_ide_0 + 0.238 * signalvector_x_cat(1);
%for n = 2: (L + N - 1)
%    output_vector_ide(n) = 0.724 * output_vector_ide(n - 1) + 0.238 * signalvector_x_cat(n);
%end
output_vector_ide = zeros(L, 1);
output_vector_ide(1) = 0.724 * s_ide_0 + 0.238 * signalvector_x(1);
for n = 2: L
    output_vector_ide(n) = 0.724 * output_vector_ide(n - 1) + 0.238 * signalvector_x(n);
end
output_ide = output_vector_ide;

%h_ide_cat = vertcat(h_ide, zeros((length(signalvector_x) - length(h_ide)), 1));
%
%output_vector_ide = conv(signalvector_x, h_ide_cat); % test point

%for n = 1: L
%    output_ide(n) = output_vector_ide(n); % L samples cutoff
%end

savefile = 'WienerFiltering.mat';
save(savefile, 'output_vector_fir', 'output_fir', 'output_vector_ide', 'output_ide');

clear;

WienerFilter_MSE();
%WienerFilter_Plot();

clear;