This is actually the third step of the problem. It's preceded by these questions that I'm sure are supposed to lead me to solution.
$n = pq$, p and q distinct odd primes
First I'm supposed to show that $y^{\frac{1}{2}\phi(n)} \equiv 1 \pmod{ p}$ and that $y^{\frac{1}{2}\phi(n)} \equiv 1 \pmod{ q}$. $gcd(y, n) = 1$
Since $\phi(n) = (p-1)(q-1)$, I showed that $y^{\frac{1}{2}(p-1)(q-1)} \equiv [y^{\frac{1}{2}(q-1)}]$$^{(p-1)} \equiv 1 \pmod{ p}$ (by Fermat's Theorem) and did basically the same thing for the other equivalence relation. Does Fermat's theorem apply in this case? I'm hoping that it's safe to assume $gcd(y^{\frac{1}{2}(q-1)}, p-1) =1$
Then I was supposed to show that $y^{\frac12\phi(n)} \equiv 1 \pmod{ n}$ from the previous steps, but I said that $y^{\frac12\phi(n)} \equiv [y^{\frac12}] ^{\phi(n)} \equiv 1 \pmod{ n}$ by Euler's Theorem (again making a similar assumption)
Then comes the question I posted in the title: Proving that if $ed ≡ 1 \pmod{\frac12 φ(n)} $, then $y^{ed} ≡ y \pmod{ n}.$
I'm not really seeing the connection between these different statements which makes me think that I did the first ones wrong. I'm thrown off by the $\frac12$ because otherwise it's easy to see that $y^{ed} ≡ y \pmod{ n}. \implies ed ≡ 1 \pmod{ φ(n)} $. I'm not seeing how to prove that it still works with the $\frac12$.
Any advice or anyone see where I did something wrong? Thanks!
Since $y$ is prime to $n$ (and $p$), so is $y^{\frac12(q-1)}$ amd indeed little Fermat tells us that $y^{\frac12\phi(n)}\equiv 1\pmod p$. By symmetry, also $y^{\frac12\phi(n)}\equiv 1\pmod q$. Then by the Chinese Remainder Theorem, $y^{\frac12\phi(n)}\equiv a\pmod n$. So finally if $ed=k\cdot\frac12\phi(n)+1$ we find $y^{ed}=\left(y^{\frac12\phi(n)}\right)^k\cdot y\equiv y\pmod n$.