Considering the input limitations of the CK37 it would seem to have been designed to operate in quarter(ish) hemispheres. Sweden, as it happens is comfortably contained within the first 99 degrees east in the north hemisphere, which is probably why.
Now the question is, was the CK37 hardcoded to only work within that constraint, or was there a way to tweak those assumptions that it makes?
It might depend on what underlying data type the CK37 used for it's coordinates.
Does anyone know how this was dealt with in the real life Viggen?
Perhaps all you need is an offset setting in the CK37, instead of having flags for N/S and W/E. Actually, that's probably what I'd have done if I was an engineer at that time, haha. That would also explain a little how NTTR could work, but I still wonder how that stacks up to reality.
As an anecdote, I discovered this coordinate confusion as I tried out the RB-15 for the first time. I must have tried to enter the coordinates for the missile 5-6 times, confirming that the numbers read back alright and everything, but when I approached the target to fire there was always a red light and no symbology where I expected. Pretty confusing but eventually I saw what was going on. Luckily, leaving Bx6-9 undefined and radar fixing Bx8 gives you a standard layout and you can fix each missile markpoint after that. Less pre-planned as I wanted but at least no show-stopper.