Is it possible to solve system of linear equation in noisy case with cvx?(when number of variables=number of equations)
Thanks for any help
You need to describe exactly what you want to do. What are you talking about?
If what you want to do is solve a linear least squares problem in which the number of variables equals the number of equations (data points), the answer is yes. See, for example, http://cvxr.com/cvx/doc/quickstart.html#least-squares .
If you want to exactly solve a system of linear equations (whether “noisy” or not), then CVX can be used to attempt to solve it, and should be able to find a solution if the matrix is of full rank. But you don’t really need CVX for this, you can just use MATLAB’s backslash operator \. Actually, you can use MATLAB’s backslash operator to solve an unconstrained linear least squares problem as well.