How to express the complex objective function in cvx?

1

The objective function is related to summation, and the problem is as follows:
%E5%9B%BE%E7%89%87
where %E5%9B%BE%E7%89%87,

%E5%9B%BE%E7%89%87
图片.png941×488 37.2 KB
.and ‘m’ represents the m-th iteration. Similarly, ‘m-1’ denots the (m-1)-th iteration.

In the objective function , N is large, i.e., 100~1000. In addition, I_L is an unity matrix (L=2), N_B=5, N_E=3. Q_B is a matrix with size LLN, and Q_E is also a matrix with size LLN. P_u and P_a are diagonal matrices with size LLN.

I think it is very difficult for me to express the objective function when N is large, and I don’t think it could be expressed with for-loop if N is very large.

So could somebody please help me that? Thanks in advance.

2

Perhaps you should show a complete code, using for loops. Try running it on s small-scale problem first in order to make sure the code is correct. Then try a larger problem. if it is too slow on the larger problem, then perhaps someone can offer suggestions to speed up your code.

For loops are always allowed, but they may not result in the fastest code. How big is the N for the problems you really want to solve?