i want to draw this DOP figure with these specific parametres

Question

Ayoub 2023-9-17

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2022262-i-want-to-draw-this-dop-figure-with-these-specific-parametres

评论： Star Strider 2023-9-17

采纳的回答： Star Strider

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Star Strider 2023-9-17

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2022262-i-want-to-draw-this-dop-figure-with-these-specific-parametres#answer_1312187

在 MATLAB Online 中打开

This probably represents a multi-exponential decay, because a single exponential doesn’t faithfully reproduce it.

I would need the signal itself rather than an image of it, to estimate all its parameters.

Fs = 1E+9;

L = 2E-6;

t = linspace(0, L*Fs, L*Fs+1)/Fs;

k = log(0.6)/0.2E-6;

s = -4*(exp(k*t) .* sin(2*pi*2.5/5E-7*t));

figure

plot(t, s.*heaviside(t-0.5E-6))

grid

ylim([-1 1.2])

.

2 个评论
显示无隐藏无

Ayoub 2023-9-17

the real signal contains noise and the aim is to remove the noise and compare it with the signal in the previous image.

the noising signal looks like this:

the parametres extracted were the folows:

The average rise time : 30 ns

decay time : 35ns

pulse width : 65ns

it is also mentioned that the equation used was this:

where A is the peak value and is assumed to be 1, t1 and t2 are damping

coefficients that determine the pulse waveform, and fc is the oscillatory frequency of DOP

Star Strider 2023-9-17

O.K.

In that instance, you need to calculate the fft of the original signal to understand where the noise is with respect to the signal peak (that should be centred about

).

A relatively easy way to do this is to either calculate the fft itself, or use the pspectrum function to estimate it.

Then use the bandpass function to filter it, eliminating the low-frequency baseline variation and the high-frequency noise. For bandpass and its friends, use the 'ImpulseResponse','iir' name-value pair to design an efficient elliptic filter to filter it. It may be necessary to threshold it, so initially eliminating the (relatively) constant section from 0 to 0.6 µs before filtering it. Another way to approach it would be to first use a lowpass filter to eliminate the high-frequency noise, and then a highpass filter to eliminate the baseline variation.

That is how I would approach it, at least initially, changing as necessary to get the most optimal result.

You can then experiment with the approach in Curve fitting to a sinusoidal function to estimate the parameters. It would be necessary to change the ‘fit’ function in that code to match the

function, so you get the appropriate information from it.

请先登录，再进行评论。