CVX optimization of semidefinite relaxation (SDR) with complex matrix

I need to optimize the following semidefinite relaxation problem

Here is my code with CVX tool

where N is the size of matrix W. There is an error say "Expressions in objective functions must be real." I don't know how to solve it.

Answer 1

[Shivani Dixit]

Hello,

From the code provided it seems that you are calculating the traces of a complex valued matrix and then comparing it with some value. To my understanding such an operation is not meaningful since complex comparison is not valid.

If this is the case you need to ensure that the values are real before comparison.

If this is not the case please do let me know the values that your matrix is using to figure out the issue.