“…By performing an analysis of the noise pedestals present in these harmonics, the amplitude jitter and timing jitter [2, 3, 5, 6] of the ML pulses can be identified, and then evaluated via integrations of the noise power in the pedestals. A comparison of these two approaches, given in [6, 8], shows that they provide consistent results if the downconversion process used in the first approach does not incur excess noise, and a sufficiently wide measurement range for Δ f is used to characterise L ( f ). For the second approach, wideband RFSA and photodetectors (PDs) must be available to accurately measure the noise pedestals for at least a few harmonics of the ML pulses.…”