I'm reading a set of notes but I don't understand the following concept. We have a long exact sequence
$\cdots \rightarrow 0 \rightarrow H_n(\mathbb{S}^m) \rightarrow H_{n-1}(\mathbb{S}^m) \rightarrow 0 \rightarrow \cdots$
Apparently this implies that $H_n(\mathbb{S}^m)$ is isomorphic to $H_{n-1}(\mathbb{S}^m)$. Why is this true? I tried to prove it using exactness and rank nullity but all I could show is that rank$(H_n(\mathbb{S}^m)) = $ rank$(H_{n-1}(\mathbb{S}^m))$
It might be clearer if you actually name the maps. You have a short exact sequence within the long exact sequence $$ 0\xrightarrow{f}H_n(S^m)\xrightarrow{g} H_{n-1}(S^m)\xrightarrow{h} 0 $$ where $f$ and $h$ are zero maps. Then by exactness, $0=\operatorname{im}(f)=\ker g$, so $g$ is injective. Also, $\operatorname{im}(g)=\ker h=H_{n-1}(S^m)$ since $h\equiv 0$. Thus $g$ is surjective, hence an isomorphism.
There is no need to look at the ranks of the groups.