Lebesgue's criterion for Riemann-integrability - a maximal function approach possible?

169 Views Asked by At

Lebesgue's criterion for Riemann-integrablity:

A function $f\colon[a,b]\to\mathbb R$ is Riemann-integrable if and only if it is bounded and continuous almost everywhere.

There are several proofs. I know one using du Bois-Reymond's criterion: a bounded function is Riemann-integrable if and only if for each $\epsilon>0$, there exists a positive number $\delta>0$ and a partition such that the total length of intervals on which the oscillation $>\epsilon$ is less than $\delta$. However, I would like to see whether we have a proof using the idea of maximal functions.

Extend the function from $[a,b]$ to $\mathbb R$ by $0$, we can reformulate the continuity as $\lim_{h\to0}(f(x+h)-f(x))=0$ for a.e. $x\in\mathbb R$.

We have seen many examples of convergence a.e. of which the proof takes advantage of a maximal function. For example, Lebesgue differentiation theorem follows from the fact that the Hardy-Littlewood maximal function is a bounded sublinear operator from $L^1$ to $L^{1,\infty}$. Similarly, Carleson's theorem follows from, say, the boundedness of the maximal Carleson operator.

I wonder whether we can prove the Lebesgue's criterion in a similar fashion.

1

There are 1 best solutions below

0
On

Of course even if the question were more precisely formulated it's hard to imagine how one might prove the answer is no. But offhand it's hard to see how a proof via a maximal function is going to work. Because, by analogy with the results you mention that are proved using maximal functions, it seems like the relevant maximal function here is just $$Mf(x)=\sup_h|f(x+h)|,$$but that makes $Mf$ constant.