Matlab Coherence function
조회 수: 6 (최근 30일)
이전 댓글 표시
The wavelet tool box I recently purchased does not seem to have the kind of coherence function that I need. This may be a lack of understanding on my part but what I was looking for was a function that is described as a magnitude squared coherence function which produces a single vector that varies from 0 to 1. It is not at all clear to me how I can get this out put from the function called wcoher that is in my just got it tool box. Any advice would be appreciated.
Thanks
댓글 수: 1
Jakub
2013년 4월 12일
Wcoher function gives you wavelet coherence. It's function of time(samples) and scales. Scales represent frequency(see scal2frq function). Input of wcoher are two vectors. You obtain information about coherence for each pair of samples and for each scale(frequency).
채택된 답변
추가 답변 (2개)
Tahani
2013년 8월 15일
Is there any way to display the coherence vector as a coloured image. for example blue colour refers to high coherence and red refers to low coherence. in addition to the coherence values between those significant markers.
댓글 수: 0
Youssef Khmou
2013년 4월 12일
hi Don
try this function :
function [MSC]=coherence_MVDR(x1,x2,L,K)
%%This program computes the coherence function between 2 signals
%%x1 and x2 with the MVDR method.
%%This algorithm is based on the paper by the same authors:
%%J. Benesty, J. Chen, and Y. Huang, "A generalized MVDR spectrum,"
%%IEEE Signal Processing letters, vol. 12, pp. 827-830, Dec. 2005.
%%x1, first signal vector of length n
%%x2, second signal vector of length n
%%L is the length of MVDR filter or window length
%%K is the resolution (the higher K, the better the resolution)
%initialization
xx1 = zeros(L,1);
xx2 = zeros(L,1);
r11 = zeros(L,1);
r22 = zeros(L,1);
r12 = zeros(L,1);
r21 = zeros(L,1);
%construction of the Fourier Matrix
F = zeros(L,K);
l = (0:L-1)';
f = exp(2*pi*l*j/K);
for k = 0:K-1
F(:,k+1) = f.^k;
end
F = F/sqrt(L);
%number of samples, equal to the lenght of x1 and x2
n = length(x1);
for i = 1:n
xx1 = [x1(i);xx1(1:L-1)];
xx2 = [x2(i);xx2(1:L-1)];
r11 = r11 + xx1*conj(xx1(1));
r22 = r22 + xx2*conj(xx2(1));
r12 = r12 + xx1*conj(xx2(1));
r21 = r21 + xx2*conj(xx1(1));
end
r11 = r11/n;
r22 = r22/n;
r12 = r12/n;
r21 = r21/n;
%
R11 = toeplitz(r11);
R22 = toeplitz(r22);
R12 = toeplitz(r12,conj(r21));
%
%for regularization
Dt1 = 0.01*r11(1)*diag(diag(ones(L)));
Dt2 = 0.01*r22(1)*diag(diag(ones(L)));
%
Ri11 = inv(R11 + Dt1);
Ri22 = inv(R22 + Dt2);
Rn12 = Ri11*R12*Ri22;
%
Si11 = real(diag(F'*Ri11*F));
Si22 = real(diag(F'*Ri22*F));
S12 = diag(F'*Rn12*F);
%
%Magnitude squared coherence function
MSC = real(S12.*conj(S12))./(Si11.*Si22);
댓글 수: 0
참고 항목
카테고리
Help Center 및 File Exchange에서 Continuous Wavelet Transforms에 대해 자세히 알아보기
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!