"Bruno Luong" <email@example.com> wrote in message <firstname.lastname@example.org>... > > > I suggest you see if ?fmincon? can do what you want. I don't have much experience with ?fmincon? so I won't attempt an example. > > No, fmincon is suitable not for binary optimizaton. > > > > > As to the solution being unique, that likely depends on the structure of A. You?ll have to experiment with it. > > No again; a reasoning that is true for real variables is no longer true for discrete variables. > > Bruno
I got the impression from the OP that A, x, and b were all _complex_ and not necessarily discrete or binary. The constraint is that the _magnitude_ of the individual elements of the complex x-vector, abs(x(k)) = 1 for all k. Thus my reply.
BTW, you?ve posted a number of elegant programming solutions, some of which I've found extremely useful. I suggest you create a function from your post in thread #319426 and upload it to FEX. I've seen the same question posted a few times since, so there seems a need for it.