Canonical: https://knowlpedia.com/harmonic-analysis/hilbert-transform/

# Hilbert transform

The singular integral operator with principal-value kernel 1/(pi x), equivalently the Fourier multiplier -i sign(xi).

For \(f\) in the [Schwartz space](https://knowlpedia.com/functional-analysis/schwartz-space/) on
\(\mathbb R\), its **Hilbert transform**
is
\[
Hf(x)=\frac1\pi\operatorname{p.v.}\!\int_{\mathbb R}
\frac{f(x-t)}{t}\,dt.
\]
The operator extends uniquely to a bounded [linear map](https://knowlpedia.com/linear-algebra/linear-map/) on \(L^2(\mathbb R)\).

## Fourier multiplier

With the Fourier convention \(e^{-2\pi i x\xi}\),
\[
\widehat{Hf}(\xi)=-i\,\operatorname{sgn}(\xi)\widehat f(\xi).
\]
This identity proves \(L^2\)-boundedness using the
[Plancherel theorem](https://knowlpedia.com/harmonic-analysis/plancherel-theorem-lca/).

## Harmonic conjugates

If \(u\) is the [Poisson
extension](https://knowlpedia.com/harmonic-analysis/poisson-extension-upper-half-plane/) of boundary data \(f\), then \(Hf\), up to the chosen sign, is the
boundary value of a harmonic conjugate. Consequently the
[Dirichlet-to-Neumann
operator](https://knowlpedia.com/harmonic-analysis/dirichlet-to-neumann-upper-half-plane/) is \(f\mapsto H[-f']\).

## References

1. Loukas Grafakos, *Classical Fourier Analysis*, 3rd ed., Springer, 2014. [DOI record](https://doi.org/10.1007/978-1-4939-1194-3).
