How to fit MLE function (MATLAB) to multivariable distribution estimation ?

I am trying to estimate parameters of custom pdf distribution , I am not sure if I could use MLE function in matlab or I should use optimisatio toolbox ( I tried using global search with fmincon but the solution does not converge to the true parameters). I have two custom pdfs I am using to estimate parameters (data is z).

1) first model pdf :

f(x)= (1-c1)*lam*exp(-lam*z)+c1*lam*u*(exp(-u*z)-exp(-lam*z))/(u-lam);

for the first model , I am trying to estimate lam and u and c1.

2) second model pdf:

f(x)=f1(x)*f2(x)*f3(x)= lam*exp(-lam*z)*u*exp(-u*z)*g*exp(-g*z);

so for the second model: I am trying to estimate lam, g and u

I want to solve the above problem but not sure about the correct approach

Also, I read the mle documentation and I couldnt understand censor in the mle function ? What does censor mean in the context of my data (if I am using random simulation and real data for estimation)