How to calculate the correct deconvoluition?

1 次查看（过去 30 天）

显示更早的评论

A.O. 2022-8-22

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1783190-how-to-calculate-the-correct-deconvoluition

在 MATLAB Online 中打开

We have 'step type' model data 'M'.

We calculated the convolution of the model data ('C') using the transfer function h(t).

Why can't get correct solution when we deconvolute the convoluted data ('C')?

r equals to C using this code.

If calculated correctly, r equals to M.

Please tell me the correct way to use the function 'deconv'.

close all
%% Step1　Definition of transfer function h(t)
% Time range
n1 = 100;
t = [1:100]';
% Input A and B
A = 5.12
B = 24.8
h = sqrt(A/(pi*(B^2)))*exp(-1/(4*A))*exp(-1*A*((log(t/B).^2)))
%% Step2　Model data 'M'（Solution of deconvolution）
% Model data
M = zeros(1,n1);
M = M';
M(1) = 0;
M(2) = 1;
M(3) = 2;
M(4) = 3;
M(5) = 4;
M(6) = 5;
M(7) = 4;
M(8) = 3;
M(9) = 2;
M(10) = 1;
M(11) = 0;
%% Step3　Model data 'C'（Convolution of 'M'）
% Convoluted data
C = conv(M, h, "full");
C = C(1:n1);
%% Step4　Calcurate deconvolution (Deconvolution of 'C')
% Calcurate deconvolution of 'C'
[q,r] = deconv(C,h);
plot(r)