I have an equation which I need to get its (least) fixed point for. Generally, we have:
x_{k+1} = f (x_{k}) # fixed point iteration
f: N -> N # N is the set of natural numbers
Is it possible to use the same fixed point theorem to prove that:
- The function admits at least one fixed point
- Finding the fixed point is guaranteed via an iterative algorithm
if the domains are sets of integers, and if so is there a source in literature?
Note: The iterative algorithm just assumes an initial value for x and computes the next value until the values are equal.
If needed, the functions which I refer to come from worst case response time of scheduling non-preemptive fixed-priority tasks in a task set.
# ceil() is the ceiling function
# equation one
w = a + b(1 + ceil(w/c))
where w is the variable and a, b, c are constants.
The form of the equation is a dummified version from this paper of Theorem 15. But the form should suffice anyway.