Objective In recent years, single-frequency lasers have been widely used in the fields of fiber communication, lidar, and fiber sensors due to their narrow linewidth and good stability. However, in high-precision applications, such as precise spectral measurement, gravitational-wave detection, and high-precision frequency transfer, noise characteristics are also important parameters and worth further optimization. High-precision measurement of single-frequency laser noise is the basis of laser noise analysis. So far, the current noise measurement method usually evaluates noise performance in a specific frequency band and hence it becomes difficult to cover the whole mHz to MHz frequency range. To realize broadband noise measurement, it is necessary to adopt segmented measurement schemes and use spectrum splicing techniques. However, there is still a lack of standard measurement schemes for obtaining the broadband noise spectrum. In this paper, a standard measurement technique for mHz to MHz broadband laser noise is reported.

Methods Among the various methods of frequency noise measurement, the optical heterodyne beat-frequency method is suitable for measuring low frequency noise, but the maximum measurement range is limited, and the correlation delay self-heterodyne measurement technique based on a fiber-type interferometer is quite suitable for high-frequency noise measurements above 1Hz. Therefore, a correlation delay self-heterodyne measurement system based on a fiber-type Michelson interferometer [Fig. 1(a)] is designed and developed for measuring high-frequency noise by suppressing environmental noise through passive control techniques such as sound insulation, vibration isolation, and temperature control, where the 10Hz--1MHz spectral range can be successfully measured. The optical heterodyne beat-frequency method [Fig. 1(b)] is used to measure low frequency noise by the beating of two self-developed DBR fiber lasers with similar performance, where a reference laser with calibrated frequency noise can be obtained. The laser to be tested is beating against the reference laser to achieve frequency noise measurements in the low frequency range of 1 mHz--100Hz. By using the direct average method, the spectra measured by these two methods can be spliced together smoothly, and the frequency noise measurement in the frequency range of 1 mHz--1 MHz can be ultimately realized.

For measuring relative intensity noise, the direct measurement method based on the electric spectrum analyzer is used in the high-frequency range, and the digital measurement method based on the digital multimeter and FFT analyzer is used for frequencies lower than 10kHz. Combining these two noise measurement techniques and using the gradual in and out spectrum splicing method, the relative intensity noise spectrum in the frequency range of 1 mHz--50MHz is obtained.

Results and Discussions Using the measurement schemes mentioned above, we evaluate the noise spectral characteristics of a single-frequency DBR fiber laser from NP photonics. We measure the frequency noise spectra with the developed correlation delay self-heterodyne method (Fig. 2) and beat frequency method (Fig. 3), respectively. And we measure the RIN spectra by the ESA [Fig. 5(a)] and FFT analyzer [Fig. 5(b)]. The measurement results are compared with the data provided by the supplier and the results reported in the related literature. At the same time, through mutual verification between different schemes, the reliability and accuracy of our measurement systems are verified. The frequency noise spectra (Fig. 4) and RIN spectra (Fig. 6) in the frequency range of mHz to MHz after spectrum splicing are obtained. It can be seen that the overlapping area in the spectrum shows a smooth curve without splicing traces. In addition, we also evaluate the broadband noise characteristics of our self-developed 1064 nm and 1560 nm single-frequency DBR fiber lasers. Compared with a commercial NP laser, our self-developed 1560 nm single-frequency DBR fiber laser shows better noise performance over a wide frequency range.

Conclusions A standard technique for measuring single-frequency laser frequency and intensity noise in an ultra-wide Fourier-frequency range from mHz to MHz is presented in this paper. We successfully measure both the frequency noise and intensity noise spectral characteristics of single-frequency lasers in a Fourier-frequency range of mHz to MHz, using two established measurement systems in conjunction, namely, a correlation delay self-heterodyne frequency noise measurement system based on a fiber-type Michelson interferometer and an optical heterodyne beat-note measurement system with calibration function, combined with common analysis instrument. The accuracy of our measurement results has been verified. This presented method may be used for laser noise evaluation in various applications such as gravitational-wave detection and precision measurement.