如何确定FFT时延时间？
问题描述：情况1：信号1在信号2之前，可准确定位时延，即程序中Nn<N1，可准确求出时延结果为N1-Nn
情况2：信号1在信号2之后，时延应是负值a，结果为正值，是信号长度L与真正时延a之和（在程序中为Nn>N1，真正时延结果为 N1 - Nn，但程序结果为L+N1—Nn）。若单纯的减去信号长度L，则当信号2在信号1之后，且时延为L+a时，这两种情况会出现混绕。
请问各位大侠，如何解决这种问题。下面是程序部分。
%% 广义自相关法时延估计函数可行性测试部分
clc; clear all ;close all;
wn =rand(10240,1);
phi = ;
signal = 0;
for i = 1:5
signal = signal + 1* sin(2*i*pi*phi);%.*exp(-(2*i*pi*phi));
end
signal = signal/max(abs(signal));%信号归一化
temp = zeros(10240,1);
Nn = 1;
temp (Nn:length(signal)+Nn-1)= signal;
signal1 = wn +temp;
% plot(signal1);hold on;
N1 = 1001 ;
temp1 = zeros(10240,1);
temp1(N1:N1-1+length(signal))= signal;
wn1 =rand(10240,1);
signal2 = wn1 +temp1;
plot(signal2,'r');
%%
= size(signal1);
D= GCC(signal1,signal2,'scot');
msgbox(['d = ',num2str(N1-Nn)'D1= ',num2str(D)])%,'D2= ',num2str(D2)]);
子函数 GCC.m
function = GCC(Sig1,Sig2,m,Fs,frame,N)
if nargin<2
error('Wrong number of input arguments');
elseif nargin == 2
m = 'roth';
end
Pxx = (abs(fft(Sig1))).^2;
Pyy = (abs(fft(Sig2))).^2;
Pxy = conj(fft(Sig1)).*fft(Sig2);
N = size(Pxy,1);
% define pre-whitening filter
switch lower(m)
case 'unfiltered'
% Unfiltered Cross-Correlation (UCC)
% -----------------------------------------------------------------
% this processor doesn't filter the cross-power applying a sort of
% time cross-correlation.
W = ones(N,1);
case 'roth'
% Roth filter
% -----------------------------------------------------------------
% this processor suppress frequency regions where the noise is
% large than signals.
if size(Pxx)~=size(Pxy)
error('Roth filter: power spectra size must be the same');
end
W = ones(N,1);
nonzero = find(Pxx(:,k));
W(nonzero,k) = 1 ./ Pxx(nonzero,k);
case 'scot'
% Smoothed Coherence Transform (SCOT)
% -----------------------------------------------------------------
% this processor exhibits the same spreading as the Roth processor.
if size(Pxx)~=size(Pyy) | size(Pxx)~=size(Pxy) | size(Pyy)~=size(Pxy)
error('SCOT filter: power spectra size must be the same');
end
W = ones(N,1);
W_Den = (Pxx .* Pyy) .^ 0.5;
nonzero = find(W_Den(:,k));
W(nonzero,k) = 1 ./ W_Den(nonzero,k);
otherwise error('Method not defined...');
end
% apply the filter
R = Pxy .* W;
% estimate the generalized cross-correlation (GCC)
G = abs(ifft(R));
D = find(G==max(G),1,'first') ;
[ 本帖最后由 sanzhxl 于 2009-7-23 09:46 编辑 ]
页:
[1]