how to add a normal distribution to a figure?

Hello

I want to creat this figure (regenrate paper)

now I got this one

how to add some normal distribution to my figure?

[Adam Danz]

It's likely that there are details in the paper that describe the functions used to produce the figure. Perhaps the curves are fits or idealized data. In both cases, it would be described in the methods if the paper is any good.

There are lots of way to produce curves (my answer shows 2) and with enough trial & error, I'm sure the figure could be approximatly recreated. But is that the goal, to draw a picture? Or is the goal to recreate the methodology in the paper which would require understanding and implementing the the underlying functions?

[masoud avaznejad]

I want to create some normal distribution with specific area between them and X axis

[Adam Danz]

Normal distributions (I'll call them Gaussians) are defined by two parameters, as minimum: center and width (mean and std). Given an area, there are lots of (infinite?) combinations of parameters that could result in that area.

Answer 1

[Adam Danz]

You can use this fully parameterized guassian function explained here

https://www.mathworks.com/matlabcentral/answers/370386-how-to-plot-a-gaussian-1d-in-matlab#answer_465773

It can be simplified if you don't need the vertical offset or amp terms.

gaus = @(x,mu,sig,amp,vo)amp*exp(-(((x-mu).^2)/(2*sig.^2)))+vo;

Alternatively, since the curves you shared show square waves and gaussians, you might be looking for a higher-order Gaussian or super-Gaussian function.

supGaus = @(x, sigma, center, amplitude, k) amplitude .* exp(-((x-center).^2/(2*sigma.^2)).^k); 

• x: a vector of x-values
• sigma: width parameter
• center: defines center of curve or the mean of the guassian
• amplitude: defines the height of the curve
• k: the 'shape' paramter which determines the steepness of falloff and how flat the top of the curve is. When k=1 the curve will be Gaussian. When k=inf the curve will be a step function.

x = 400:5:2100; 
center = 1260; 
sigma = 250; 
amplitude = 60; 
% showing results for various k=values defined in legend