Combined with current testing technology and space-based gravitational wave detection requirements, relative intensity noise (RIN) testing must cover the frequency range of 0.1 mHz?5 GHz. Currently, low-noise spectrum analyzer is used for RIN testing in the higher frequency band of 50 kHz?5 GHz, and relevant theories and testing methods are relatively mature. OEWaves of the USA SYCATUS of Japan and Shanghai Institute of Optical Machinery have launched corresponding standard test instruments. However, the current test methods in the low-frequency band are limited in the test band or have high background noise, which cannot fully meet the requirements of RIN low frequency band test and evaluation of laser light source for space-based gravitational wave detection. It is necessary to develop the low background noise measurement technology and complete and accurate evaluation standard of all low frequency band.
In this study, the low-frequency RIN within 0.1 mHz?100 kHz is completely tested and characterized, and the background noise of the test system is reduced to form a standardized test system and test algorithm. First, based on low-noise photodetector, high-precision digital multimeter, Labview control data acquisition, and data processing algorithm programming, the test characterization of laser RIN in the frequency band of 0.1 mHz?0.5 Hz was realized. In the time domain, the high-precision acquisition was conducted using the Labview software to control DMM. The fast Fourier transform (FFT) algorithm was used to analyze the noise characteristics of the collected data in the frequency domain. The smoothing function of different resolutions was used in the calculation of RIN to ensure that the test results in the low-frequency band are not true, while the serious "trailing" phenomenon in the high-frequency band was reduced. Besides, frequency domain analysis can be performed immediately upon the completion of the time domain collection, and the data can be stored in real time. Combined with the data of different sampling time, the accuracy of the very-low-frequency test results was verified. Second, FFT spectrum analyzer (SR770, Stanford Research Systems) was used to test the RIN of laser in the frequency band of 1 mHz?100 kHz. By adding low noise amplifier (LNA) into the test system, the background noise in the frequency band of 1 mHz?1 Hz was effectively reduced. The testing capacity was reduced by 18 dB. Finally, the consistency of the test results of the two test methods in the overlapping frequency band was compared to verify the uniformity and accuracy of the two test results. Finally, the low-background-noise RIN test band was expanded to 0.1 mHz?100 kHz. The RIN noise measurement system has the advantages of wide coverage of low-frequency band, high precision, and high accuracy. It can provide a standardized measurement means for the relative intensity noise of laser in space gravitational wave detection and can also be applied to other low-frequency precision measurement applications of laser light source noise assessment.
According to the sampling time listed in Table 3, data segments ranging from 10000 s to 8000 s are captured to calculate the laser very-low-frequency RIN, respectively, and the results are shown in Fig. 7. It can be seen that the high-frequency "tail" after Smooth piecewise smoothing algorithm is only 2 dB; in the range of 0.04 Hz to 0.5 Hz, different data lengths have little influence on RIN results. However, in the frequency range of 0.1 mHz?0.04 Hz, the sampling time of 8000 s is significantly different from that of 10000 s, and more noise information can be detected at 80000 s. Moreover, Fig. 7 shows that the curves with the sampling time of 10000 s and 2000 s have poor coincidence compared with other long periods. This is because the short test time leads to fewer data points in the frequency band of 0.1 mHz?0.04 Hz, resulting in decreased accuracy. In combination with the abovementioned and theoretical analysis, it can be seen that the longer the sampling time, the more accurate the test results.
Figure 8(a) shows the RIN test results of two kinds of high-precision DMM and FFT spectrum analyzer simultaneously. Their test frequency bands cover 0.1 mHz?0.5 Hz and 1 mHz?100 kHz, respectively. As can be seen from Fig. 8(a), the two maintain a good consistency within 1 mHz?0.5 Hz in the overlapping frequency band, which on the one hand verifies the correctness of the test results. On the other hand, a complete test of RIN characteristics in the frequency band of 0.1 mHz?100 kHz can be completed by splicing the noise spectrum of the two test results. Figure 8(b) shows the complete relative intensity noise spectrum of the low-frequency band of 0.1 mHz?100 kHz obtained after splicing.
The test technique in this paper is applied to test the RIN of different types of lasers, and the characteristics of laser RIN in the low-frequency band are obtained to guide the development and optimization design of the laser and the parameter performance of the application system.
Based on the strict demand for laser noise in space-based gravitational wave detection, we complete the establishment of low-background relative intensity noise test characterization system in the low-frequency band, background noise up to -99 dBc/Hz@0.1 mHz, -165 dBc/Hz@100 kHz. This technology converts the optical signal of the laser into an electrical signal based on low-noise photodetector and performs the complete and accurate characterization of intensity noise in the range of 0.1 mHz?100 kHz through the combination of high-precision DMM, FFT spectrum analyzer, and other test means. The RIN of four typical lasers is tested and analyzed. The main noise characteristics of each laser and the subsequent available intensity noise suppression technology are obtained, and according to the noise performance of the self-developed NPRO laser, the direction of improving the relative intensity noise in the very-low-frequency band is proposed in the next stage. The relative laser intensity noise characterization test can provide accurate and unified evaluation method for laser source noise level in space gravitational wave detection and provide reference for laser source noise suppression.
