Fig. 1: Principles and experimental realization.
a Conceptual schematic showing the laser under test (LUT) combined with a probe. After the long-pass-filtered (LPF) optical signal is measured on a photodetector (PD), a low-noise amplifier (LNA) followed by a low-pass filter (LPF) are used for signal conditioning prior to sampling by an oscilloscope. b Comparison of sampling in the EFXC, and IXC. The lag increases linearly from pulse-pair to pulse-air to produce EFXC and IXC signals. c Energy diagram of the 2PA detection process. d Chart-style comparison of existing SM characterization techniques in conventional realizations with the IXC technique for a MHz-repetition-rate pulsed laser. When comparing the frame rate in dimensionless units irrespective of the laser repetition rate, it is better to use a figure-of-merit (FOM) criterion here defined as \({{{{{{{\rm{FOM}}}}}}}}=2\Delta \nu \Delta {f}_{{{{{{{{\rm{r}}}}}}}}}/{f}_{{{{{{{{\rm{r}}}}}}}}}^{2}\) (in units of Hz2/Hz2). For a Nyquist-limited sampling technique, is it lower or equal to 1, while for IXC it can reach values of 100 or more. Non-sampling techniques like DFT or IAC have a theoretical frame rate of Δfr = fr, which yields a FOM = 2Δν/fr2 approaching infinity for large optical bandwidths Δν.
