If you set the proof up as an inductive proof on one variable, and then, during the inductive step for that variable, conduct an inductive proof on the second variable, I think you'll find that everything proceeds exactly the way you would expect it.
I.e., your base case might be $n=1$, then your outermost inductive step would be to assume the statement is true for $n=k$ and attempt to prove it for $n=k+1$. Then, you would attempt to prove the statement with $n$ replaced by $k+1$ by induction on $m$. Does that make sense?
However, I should point out that the statement you're trying to prove appears to be false.