next up previous contents index
Next: Derivation of the PDE Up: -representation Previous: -representation   Contents   Index

Revision: $ P$-representation

We recall that the $ P$-representation of an operator $ \hat{\theta}$ was defined as (cf. 4.137)
$\displaystyle \hat{\theta}= \int \frac{d^2 z}{\pi}P(\hat{\theta};z)\vert z\rangle \langle z\vert.$     (69)

Remarks:

1. Other authors use a definition without the $ 1/\pi$.

2. Some books write $ P(z)$ (instead of $ P(\hat{\theta}=\rho;z)$) for the $ P$-representation of the density operator, and use the form

$\displaystyle P(z)\equiv P(z,z^*) = {\rm Tr} \left[\rho \delta(z^*-a^\dagger) \delta(z-a)\right].$     (70)

(again multiply this by $ \pi$ to get our $ P$).

3. For coherent states $ \rho=\vert z_0\rangle\langle z_0\vert$, one has $ P(z)=\pi\delta(z-z_0)$.

4. We have the Metha-formula (4.149),

$\displaystyle P(\hat{\theta};z) = e^{\vert z\vert^2} \int \frac{d^2z'}{\pi}
\langle -z'\vert \hat{\theta}\vert z'\rangle e^{\vert z'\vert^2} e^{zz'^*-z^*z'}.$     (71)

5. The $ P$-distribution can be highly singular. Example: number state.



Tobias Brandes 2004-02-18