Reputation: 681
The question is :
Assume that f, g: N → N are functions such that f(n)=O(logn) and g(n) = Ω(nlogn). Is it possible that f(n) = Ω(g(n))?
I'm thinking that it is not possible becaouse nlogn > logn, not sure if it's true nor how to proove it.
Thanks in advance !
Upvotes: 5
Views: 1000
Reputation: 18348
No, it's not possible.
Let's assume it's possible:
g(n) = Ω(nlogn)
==> there is a
such that g(n) > anlogn
for a sufficiently large n
.f(n) = Ω(g(n))
==> there is b
such that f(n) > bg(n) > banlogn
for a sufficiently large n
.c = ab
==> f(n) > cnlogn
for a sufficiently large n
==> f(n) = Ω(nlogn)
.f(n) = O(logn)
==> there is d
such that f(n) < dlogn
for a sufficiently large n
.cnlogn < f(n) < dlogn
==> cnlogn < dlogn
==> n < d/c
. That's not possible since natural number n
exists which is larger than d/c
. ==> contradiction to the initial assumption.Upvotes: 3