That's the problem, all this is very expensive and then often also slow. I'll have to dig out a filter synth software from the late 80's. Strictly DOS (had to fix the Borland error), blazingly fast, and does lots of weird architectures which I am going to need here.
Obtaining data for the system modules is tough, too. RF manufacturers are often a bit secretive. Once when they didn't want to release any internal phase and impedance data they called me back a few days later, probably to appease me because I had threatened to ditch their module. "Well, since you guys didn't release the data I measured it" ... "WHAT?"