Fig. 1. Experimental setup of frequency comparison between optical lattice clock and hydrogen maser

Fig. 2. Stability measurement of the H-maser. (a) Beat frequency ƒ_b between ⁸⁷Sr optical lattice clock and optical frequency comb (the squares represent the mean value of each measurement and error bars indicate 1σ standard deviation); (b) overlapping Allan deviation between ƒ_b and the frequency of ⁸⁷Sr optical lattice clock {circles indicate the stability calculated from the comparison data between the OLC and H-maser, the solid curve is the fitting result using the stochastic noise model, squares are the interleaved self-comparison stability of the OLC, the dotted line is the linear fitting [with a fixed slope of -0.5, showing an interleaved self-comparison stability of 6.6×10^-16 (τ/s)^-0.5], and the triangles are the stability of the hydrogen clock by comparing with the local Coordinated Universal Time (UTC-NTSC)}

Fig. 3. Simulations of H-maser random noise. (a) Three types of random noise sequences of 86400×5 s, 86400×10 s, and 86400×30 s generated by H-maser noise model; (b) calculated overlapping Allan deviation according to three types of noise sequences, and solid line indicates the H-maser noise model obtained from Fig. 2(b)

Fig. 4. H-maser measurement uncertainty as a function of the effective operation ratio