Figure 5: Block diagram of the setup for measuring the first and second moments of the OAM probability distribution carried by a paraxial light beam.

An initially Gaussian beam is converted into an OAM superposition state using a quarter waveplate (QWP) and a q-plate (QP). A polarizer (P) and a half waveplate HWP1 are then used to prepare the polarization in a linear state oriented at 45° with respect to the axes of the polarizing beam splitter PBS1. This PBS1 is the input/output port of a polarizing Sagnac interferometer (PSID)31,36, whose path is closed by mirrors M1, M2 and M3, that contains the Dove prism used to rotate the two counterpropagating beams with respect to each other. At the output of the PSID, a Babinet–Soleil compensator (BSC) is used to adjust the relative phase shift δ. The last stage is the balanced polarizing homodyne detector HD, with the axes rotated by 45° with respect to the PSID axes through the half-waveplate HWP2 (Fig. 1).