Hi,
PICs are not capable of handling the large floating point aritnmetic and are not fast enough for large systems. If you are using a one or two state filter you may use it.
You may use a simple adaptive filter that may not require significant matrix inversions.
B R M