This should be considered a research effort, assuming it will be a complete rewrite. In light of that, you should not draw down resources from the established code base to work on it.

Ignoring the above, first state the explicit requirements driving this change and let people weigh in on those. This sounds like a geeky dev itch.