Objective For some complex urban scenes under laser echo, the corresponding full-waveform echo data is inevitably mixed with various noises, which affects the extraction of effective signals. Traditional Gaussian noise reduction algorithms struggle to meet the filtering requirements of both effective and noise signals. In recent years, there has been a greater focus on the filtering effect of wavelet noise reduction, which is affected by several parameters. Therefore, in this paper, a parameter optimization wavelet noise reduction algorithm is proposed to improve the filtering effect of full-waveform data.

Methods The parameter selection of wavelet denoising is the central issue of this research. This paper employs one-dimensional wavelet denoising function (WDEN) in MATLAB to select the threshold selection criterion (TPTR), threshold usage method (SORH), threshold processing with noise change parameter (SCAL), decomposition layer (NBD), and wavelet basis function name (WNAME) five control parameters for filtering, calculates the filtering results under each parameter combination, and compares to obtain the wavelet optimal combination of control parameters. The specific steps of the parameter optimization wavelet denoising algorithm and its verification process are as follows:

1) Set five input parameter types (TPTR, SORH, SCAL, NBD, WNAME) according to the waveform characteristics, and reconstruct the echo waveform by referring to the wavelet formula (Eqs. 1~7) and the one-dimensional noise reduction function.

2) Taking the maximization of signal-to-noise ratio as the optimization goal, extract the corresponding best parameter combination of each verification waveform.

3) In the final verification, analyze the obtained experimental results according to the evaluation index of noise reduction effect.

Results and Discussions Table 3 is the optimal control parameter combination of ten waveforms, which is determined according to the maximum signal-to-noise ratio in each waveform parameter combination. The trend of these parameters is clearly visible in this table, which shows that, with the exception of the wavelet basis, the other four parameters outside the function are exactly the same. Table 4 compares the filtering effects of Gaussian, block Gaussian, and wavelet filtering under different indicators. The filtering effect in descending order is that of wavelet filtering, Gaussian filtering and block Gaussian filtering. Specific to each indicator of each waveform, wavelet filtering is far superior to the other two methods. The signal-to-noise ratio and peak signal-to-noise ratio of wavelet filtering are generally higher than the other two methods by 25 dB--35 dB, and its root mean square error and average absolute error are an order of magnitude lower. In the second-to-last row of Table 5, the average value of the ten verification waveforms is 35.84 m. The average values of the even-odd inflection point Gaussian decomposition method and the GLAH14 data file algorithm are 9.92 m and 16.42 m, respectively. They are not particularly accurate in comparison. The effective inflection point of the weak signal cannot be identified in the implementation of the even-odd inflection point Gaussian decomposition method in the urban feature scene, as shown in Fig. 7. If you want to be more precise, you must consider other methods based on the problems listed above. This article presents a component information correction method based on this, that is, a method of correcting the center position information based on the extracted Gaussian component information. The average elevation using this method is 35.47 m, which is very close to the measured average elevation of 35.84 m, and the root mean square error is only 1.02, which meets the requirements for accurate measurement of ground features, as shown in the last two columns of Table 5. The height measurement results of only the wavelet filtering under parameter optimization and the other operation unchanged methods are shown in the 5th, 7th, and 9th columns of the table. The average value of the height measurement results of the inflection point decomposition method is therefore increased from 9.92 m to 14.37 m, the root mean square error has also been reduced from 42.07 to 29.00, and the effective peak decomposition algorithm and correction algorithm therefore have limited improvements.

Conclusions In summary, the wavelet noise reduction algorithm under parameter optimization can extract the appropriate combination of many control parameters, and achieve an extremely excellent filtering effect compared to the traditional method. Data with better filtering effects can improve measurement results under less accurate methods in subsequent verification, but help for more accurate measurement methods is limited. Wavelet filtering has an effect on full-waveform data height measurement. It has a positive effect, but the focus of height measurement research should be on improving the decomposition method and adjusting the threshold control parameters in the algorithm.

Objective The solid-state slab laser has become one of the most reliable, promising and potential lasers among current high-power lasers due to its small size, lightweight and high conversion efficiency. It is commonly used in various fields, including scientific research, industry and medical treatment. High output power and good beam quality are two constant goals in the development of high-power solid-state slab lasers. As laser output power increased, the edge effects became more severe and distorted the wavefront of the solid-state slab laser output beam, resulting in a non-linear drop in laser beam quality. Several methods are present to compensate for the distortions, including the use of static compensation components, non-linear optical compensation methods, and adaptive optics. Adaptive optics is a promising method to compensate for wavefront distortion in the laser output beam. The direct slope reconstruction method is commonly used in the research of solid-state slab laser beam clean-up, and the least-squares algorithm is used to solve the system’s optimal solution.

Although adaptive optics can considerably improve the beam quality of laser output beams, some issues still need to be addressed. The least-squares reconstruction method’s criteria are to minimise the sum of slope residual squares. The adaptive optics system’s correction capability is limited by factors such as materials and cost. When the adaptive optics system can completely compensate for wavefront distortion, the least-squares reconstruction method can be used to obtain the system’s optimal solution. However, if a portion of the distortions is beyond the capability of adaptive optics system, the wavefront distortions of the laser beam cannot be fully compensated and a considerable amount of wavefront residual is still present after compensation; the minimum sum of slope residuals squares is not equivalent to the best beam quality at this time. When the solid-state slab laser operates at high gain, the wavefront distortions are very likely to exceed the adaptive optics system’s correction capability. Under these conditions, the least-squares reconstruction method cannot produce the optimal system solution.

Methods To solve the abovementioned problems, the most straightforward and effective method is to increase the number of actuators or even cascade multiple deformable mirrors with compatible wavefront sensors to improve an adaptive optics system’s inherent correction capability. However, as the number of actuators in deformable mirrors increases, their size, weight and cost also increase. Another approach is to optimise the adaptive optics system’s correction method, and a weighted least-squares reconstruction method has been proposed to improve the beam quality by assigning low weights to the uncorrectable wavefront components in the least-squares method. Unfortunately, determining the weights in a practical adaptive optics system is difficult. The edge effect remains a challenge, particularly when the number of actuators is limited because of the beam size or cost.

We proposed a novel adaptive optics correction method to further improve beam quality when wavefront distortions exceed the adaptive optics system’s correction capability. In this method, we used the idea of optimal correction in the wavefront sensor-less adaptive optics system and combined it with the traditional adaptive optics system, with the improvement of beam quality as the optimisation goal, and the optimisation algorithm is used to optimise the calibration position of the wavefront sensor according to wavefront distortions and the correction capability of the deformable mirror, and it then uses the traditional adaptive optics system for aberration compensation.

Results and Discussions We used simulation to validate the proposed method. First, we combined the two-dimensional Legendre polynomials based on the characteristics of the solid-state slab laser’s output wavefront to obtain a wavefront with severe edge distortion, and the corresponding beam quality is β=5.1 (Fig. 6). Then, the optimisation algorithm is used to find the best solution; the best beam quality that can be obtained after correction is β=1.8 (Fig. 7). Finally, the proposed method and the traditional correct method are used to compensate for the distortions (Fig. 8). After correction using the traditional method, the beam quality improves to β=3.7, whereas correction with the proposed method improves the beam quality to β=2.0, which is closer to the optimal solution. Analysing the wavefront slope distribution using a different method reveals that after processing using the proposed method, the effect of uncorrectable large distortions on adaptive optics systems is reduced (Fig. 9).

Conclusions When the correction capability of the deformable mirror is limited, the correction results obtained using different methods show that, when compared with the traditional adaptive optics system without calibration optimization the method proposed in this paper can effectively improve the correction effect of the adaptive optics system.

Objective Vision loss is caused by age-related macular degeneration because of soft drusen, diabetic macular edema and choroidal neovascular disease. Early detection and treatment of these fundus diseases have emerged as a major health concern for all countries. Professional doctors often use retinal images from optical coherence tomography (OCT) to diagnose eye diseases. However, because there are several types of retinopathy images and the lesion area is similar, manually classifying OCT retina images is a time-consuming and difficult task. With the development of artificial intelligence, researchers began classifying medical images using classic machine learning algorithms and deep learning in its branch areas, eventually progressing to the automatic classification of OCT retinal images. Several researchers are only concerned with classification accuracy and ignore the possibility of clinical application. Consequently, the network model’s parameter amount, computational complexity and floating-point operations (FLOPs) calculation amount are increasing, and the model is making inference predictions, which consumes a long time to complete. In this paper, a multi-channel, multi-scale lightweight convolutional neural network is proposed to automatically classify OCT retinal images for achieving high ophthalmic disease classification accuracy. In the future, doctors will be able to quickly view detection results in the clinic.

Methods In this study, a multi-channel OCT retinal image automatic classification deep neural network is used. The neural network model is based on the GM-OCTnet algorithm, which includes a light quantum spatial attention mechanism distinct from the convolution operator and a lightweight convolution block to replace the two modules in the original model for automatically classifying OCT retinal images. Image pre-processing, dataset division and classification using the model algorithm are the steps taken to achieve the automatic classification of the entire OCT retina. First, image cropping is performed on the collected OCT image, the blank area of the OCT image is cropped, the marginal blank area is filled and bilateral filter denoising and other pre-processing methods are used to overcome the interference of image background noise on the classification accuracy. Then, the pre-processed image is divided into three sets: training, validation, and test sets. Afterwards, the OCT images from the training set are trained using the proposed multi-channel lightweight deep neural network GM-OCTnet model algorithm. Following the test, the well-trained model is used to classify unclassified retinal images automatically. In addition, the results of this work on the automatic classification of OCT retinal images are validated by comparing the proposed model with three traditional lightweight models on the OCT data set, and different data sets are used to further validate the algorithm’s performance.

Results and Discussions Different pre-processing methods were used to process the OCT images after evaluating the quality of two different data sets (Figs. 4 and 5). The experimental results show that when the number of groups is 4, the proposed multi-channel OCT automatic retina classification network achieves an average accuracy of 96.1%, which is 2.6% higher than that of the original model GhostNet in the automatic classification of OCT retina images. Its file size is 2.64×10 ⁶smaller than that of MobileNetV3. The training and verification accuracies of the GM-OCTnet model when the grouping g=4 increase gradually with the increase of the training period and tend to stabilise, based on the relationship between the verification loss rate and the verification accuracy rate and the training loss rate and the training accuracy rate curve. When comparing the 50 training processes of different models, the proposed model is found to exhibit a higher accuracy rate than other models when the number of groups is equal to 4 and prioritises reaching the best value (Fig. 6). Overall, the model algorithm proposed in this paper has achieved high accuracy in the automatic classification of OCT retinal images. In addition, when the experimental results of two different datasets are compared, it is discovered that the automatic classification of datasets 1 and 2 achieved high classification accuracy using this algorithm (Tables 1 and 2).

Conclusions This study proposes a multi-channel, multi-scale lightweight network for automatically classifying OCT retinal images. The effect of the lightweight neural network GM-OCTnet based on the OCT image datasets in classifying and diagnosing the four types of ophthalmic conditions, i.e. choroidal neovascular disease(CNV), diabetic macular edema(DME), drusen and normal patients, were tested and evaluated. Two different datasets are used to further validate the performance of the algorithm proposed in this work. To validate the effectiveness of the GM-OCTnet model for OCT image classification, accuracy, parameter amount, calculation amount and weight file size are used as evaluation criteria. It is found the proposed OCT classification model has improved classification accuracy through experimental results. When used in the clinic, it can improve professional ophthalmologists’ diagnosis efficiency for patients with ophthalmic diseases and also reduce missed diagnosis and misdiagnosis of patients.

Objective Free-space optical communication has attracted wide attention due to its advantages, such as large information capacity, high transmission rate, strong anti-interference ability, small system size and low power consumption. Being an important part of the free-space optical communication, atmospheric laser communication is facing technical bottlenecks, such as unsatisfactory reception efficiencies and large energy jitters. Fiber coupling can effectively solve these technical bottlenecks. Coupling with optical fiber is the most common method; however, multi-mode fiber limits the usage of coherent communication reception. Using a single-mode fiber will definitely reduce the coupling efficiency under atmospheric turbulence conditions. Recently, pre-amplification of few-mode fiber or few-mode multi-core fiber has become essential to improve the reception performance of atmospheric laser communication. However, a turbulent atmosphere will have effects, such as spot distortion, arrival angle fluctuation, beam expansion and light intensity flicker. These effects seriously affect the coupling efficiency between the space light and few-mode fiber. In this study, we first studied the relationship between few-mode fiber parameters, turbulence intensity, tracking error and the coupling efficiency of space light to the few-mode fiber under the conditions of atmospheric turbulence and tracking error. Then, we established a coupling model of the space light to the few-mode fiber under the conditions of atmospheric turbulence and tracking error. Finally, we established an experimental platform under this condition to verify the coupling performance of the few-mode fiber. We hope our results can provide a technical reference for the design of the atmospheric laser communication pre-amplifiers.

Methods Based on the coupling model of space light to few-mode fiber, an approximate calculation of step-type few-mode fiber optical field was performed using the Laguerre-Gaussian (LG) distribution and LG beam as the free-space light. We obtained the optical field distribution of the LP and LG modes. According to the acquisition tracking pointing model and Von Karman turbulence spectrum model, we established the coupling model by superimposing the optical field distribution of the LP and LG modes under atmospheric turbulence and tracking error. We studied the relationship between the coupling efficiency of the few-mode fiber and the intensity of atmospheric turbulence, the angle of aiming error under different fiber V values. A self-made turbulence simulator was used to simulate the atmospheric turbulence of different intensities and experimentally study the relationship between the coupling efficiency and fiber V values under atmospheric turbulence.

Results and Discussions We obtained the relationship between the coupling efficiency and the V value of the fiber under different atmospheric turbulence intensities and different tracking error angles (Fig. 2, Fig. 5) using simulation. The coupling efficiency increases with the increase of the V value. When Cn2=0, as the V value increases to 4.85, the coupling efficiency reaches a maximum of 0.98, and any further increase in the V value decreases the coupling efficiency. This is because as the fiber V value increases, the number of spatial modes supported by the fiber also increase. Thus, a large number of modes cause crosstalk between modes. Therefore, there is an optimal V value to maximize the coupling efficiency of space light to the few-mode fiber, and as the turbulence intensity increases, the optimal V value gradually increases. We also obtained the relationship between the coupling efficiency and the V value of few-fiber under different atmospheric turbulence and tracking error angles (Figs.6--9). In the traditional free-space optical communication system, the tracking error angle is 3 μrad. Currently, under the weak turbulence of Cn2=10 ^-15 and Cn2=5×10 ^-15, the best coupling efficiencies can be obtained when the fiber V value is 5, which are 0.74 and 0.732, respectively; under the strong turbulence of Cn2=10 ^-14 and Cn2=5×10 ^-14, the best coupling efficiencies can be obtained when the fiber V value is 5.2, which are 0.715 and 0.7, respectively. The coupling experiment of space light to few-mode fibers under simulated turbulence has been completed, and the results are shown in Table 1 and Fig.13, which is consistent with the simulation results.

Conclusions We have established a coupling model of space light to the few-mode fiber under the conditions of atmospheric turbulence and tracking error, based on the acquisition tracking pointing model and the Von Karman turbulence spectrum model. This study shows the optimal V value of few-mode fiber that makes the coupling efficiency the best under the influence of atmospheric turbulence, non-linear effects and crosstalk between modes. As the V value of the fiber increases, its ability to resist tracking error also increases. Considering that the tracking error is 3 μrad, for Cn2=5×10 ^-15and Cn2=10 ^-15, the fiber V value is 5 and the best coupling efficiencies are 0.74 and 0.732; for Cn2=5×10 ^-14and Cn2=10 ^-14, the fiber V value is 5.2 and the best coupling efficiencies are 0.715 and 0.7. The coupling experiment of the space light to the few-mode fibers under simulated turbulence has been completed, the coupling efficiency of the experiment is consistent with the trend of the simulation results. The feasibility of the simulation results in practical applications is verified, providing a reference for the research of free-space optical communication.

Objective The monitoring of relative humidity is very important in the fields of agriculture, medical treatment and biochemical research, which urges scholars to develop various humidity sensors. Among many humidity sensors, optical fiber humidity sensor has become a research hotspot owing to their unique advantages such as high sensitivity, anti-electromagnetic interference, compact structure and other unique advantages. There are many kinds of optical fiber humidity sensors, and the interferometric humidity sensor has attracted wide attention because of its advantages of simple preparation and high sensitivity, among which the humidity sensor based on Mach-Zehnder interferometer (MZI) is widely used. However, the sensitivity and stability of these sensors still need to be further improved. In order to improve the sensitivity of the sensor and maintain its stability, a humidity sensor based on seven core tapered fiber is proposed and demonstrated. The sensor consists of a short section of seven core fiber between two single-mode fibers, in which the seven core fiber is fused and tapered by a hydrogen oxygen flame to form a tapered structure. Mach-Zehnder interferometer is formed by interference between the base mode of the cladding and the higher mode excited after the fiber is tapered. The effective refractive index of the cladding mode is easily affected by the external environment parameters. Therefore, the structure is very sensitive to the changes of the external environment parameters. The proposed structure is particularly suitable for the situations where high measurement sensitivity and high stability are required.

Methods Firstly, the seven core fiber with a length of about 1 cm is fused between two single-mode fibers. Then, the seven core fiber in the sensing structure is fused and tapered by a fiber taper machine (Kaipule Co. Ltd. AFBT-8000MX-H). Finally, GO film is coated on the surface of the tapered area by photothermal method. The coating process is as follows: the GO prepared by the improved Hummers method is mixed into a solution with the concentration of 1 mg/mL, and the GO solution is dropped on the surface of the fiber. In the coating process, the SLED broadband light source is used to transmit the light in the sensing structure. When the exciting light passes through the tapered fiber, a part of the light enters the cladding and generates a lot of heat in the cladding. GO molecules can be firmly adsorbed on the fiber surface by using the photothermal effect of laser, and thus the film is uniform and firm.

Results and Discussions Firstly, the wavelength scanning function of Rsoft software is used to calculate the sensor uncoated with diameter of 10 μm. The refractive index sensitivity of sensor is about 1200 nm/RIU (Fig. 4), which proves the feasibility of the experiment and provide a theoretical basis for humidity measurement. In order to provide experimental basis for subsequent humidity test, the refractive index response of samples with different diameters of uncoated GO was tested. The refractive index sensitivities of the samples with the diameter of 15 μm (s-1), 12 μm (s-2) and 10 μm (s-3) were 685 nm/RIU, 753 nm/RIU and 1123 nm/RIU, respectively (Fig. 7). Thus, the refractive index sensitivity of the sample can be greatly improved by increasing the stretching length to reduce the tapered diameter of the seven core fiber. Then, the s-4 was prepared under the same parameters as the sample with the highest sensitivity. And the GO film was coated on the surface of s-4 to prepare a humidity sensor. The experimental results show that the humidity sensitivity of sample s-4 is the highest at 1533 nm, and the maximum sensitivity is -0.0535 nm/(%RH) (Fig. 10). And the humidity sensitivity of the sensor with diameter of 14 μm (s-5) is -0.0173 nm/(%RH) (Fig. 11). Thus, the humidity sensitivity can be increased by reducing the sample diameter. In addition, we also evaluated the stability of s-4. When the relative humidity is 34.8%RH, 45.0%RH and 60.3%RH, the maximum error of the sensor wavelength is 0.03 nm, 0.04 nm and 0.04 nm, respectively, which indicates that the proposed sensor has good stability (Fig. 12).

Conclusions In summary, we propose and demonstrate a humidity sensor based on tapered seven core fiber, and the seven core fiber is fused and tapered by a hydrogen oxygen flame to form a tapered structure. The experimental results show that the refractive index sensitivity is up to 1123 nm/RIU for the sensor uncoated with taper waist diameter of 10 μm, which is consistent with the simulation results. Then the humidity sensor was fabricated by coating a layer of GO film on the surface of fiber. The maximum humidity sensitivity of the sensor was -0.0535 nm/(%RH), and the linearity was 98.5%. The sensor has the advantages of high sensitivity, simple preparation and good stability, which can be used in the field of humidity sensing.

Objective In 2015, Liu et al. used a narrow linewidth super-fluorescent fiber source as a seed to be amplified, and then achieved a 1.5 kW high-power laser output with a spectral linewidth of 0.8 nm at the highest power. It has been successfully applied in a high brightness spectral beam combining system, which lays a foundation for the application of a narrow linewidth and high power super-fluorescent source. When the narrow-linewidth laser is used as the sub source of the spectral combining system, the brightness of the spectral combining system can be effectively improved by narrowing the spectral linewidth of the laser. Therefore, it is of great significance to study a high power fiber super fluorescent source with a narrower linewidth.

Methods An all-fiber super-fluorescent source is established. Laser diode, (2+1)×1 pump-signal combiner, ytterbium doped fiber, optical isolator, optical circulator are fused as the schematic illustration shown in Fig.1. By setting the current of the laser diode lower than the stimulated radiation threshold, the laser works under a super-fluorescent state. The backward output of the broadband source is filtered by a fiber Bragg grating (FBG), and a narrow linewidth laser signal is obtained at the port 3 of the circulator. The full width at half maximum (FWHM) of this FBG is 0.16 nm, and the reflectivity is greater than 99% at the central wavelength of 1064.1 nm. In order to further improve the spectral signal-to-noise ratio of laser signal and get a narrower spectral linewidth, the laser signal passes through another filter after proper power amplification, and the FWHM reaches 0.08 nm. After two stages of pre-amplification, the power of the laser signal is increased to 15 W, which is injected into the main power amplifier stage and then amplified to 1.08 kW. The spectra for different output powers in the main power amplifier stage are shown in Fig.2 (b). A 4 km G652D passive fiber is fused between the second stage pre-amplifier (Amp2) and the third stage pre-amplifier (Amp3), and then the spectrum at 1.08 kW is narrowed as shown in Fig.3 (a). The comparison of spectral linewidth versus output power is shown in Fig.3 (b) with and without the 4 km passive fiber.

Results and Discussions By comparing the results of these two experiments, it can be found that the spectral FWHM is narrowed by 0.03 nm and the RMS linewidth is narrowed by 0.06 nm at a 1.08 kW output power by fusing a 4 km passive fiber between Amp2 and Amp3. Limited by the available pump power, the change of spectral width at a higher power has not been further verified, but according to the linear broadening law of laser spectra, this method via fusing an additional long-distance energy transfer fiber to compress the laser output spectrum will play a more obvious role in the laser at higher power. Some previse researches have shown that the time-domain characteristics of laser seed source will affect the speed of spectral broadening, but the comparison of the time-domain probability density functions (PDFs) of the laser signals with and without the 4 km fiber, shown in Fig 4(b), implies that the improvement of time stability is not very significant. In order to explore the reasons for the change of spectral broadening speed, the length of the passive fiber is reduced to 3 km, the spectral FWHM is broadened from 0.18 nm at 15 W to 0.23 nm at 1.08 kW, and the broadening slope is 0.05 pm/W. By further shorting the passive fiber to 2 km, the spectrum is broadened from 0.16 nm to 0.25 nm with a 0.09 pm/W slope. By lengthening the passive fiber to 5 km, the change of FWHM versus power is same as that for 4 km fiber. The difference is that the spectral side-mode is enhanced. One possible explanation is that the dispersion of a long-distance fiber can change the phase matching condition between the sub-wavelengths in the laser signal and weaken the intensity of the FWM effect, and then the spectral broadening speed in the process of power amplification is restrained.

Conclusions A narrowband all-fiber super-fluorescent source is established with a power of 1.08 kW after multi-stage power amplifiers. The full width at half maximum is 0.23 nm under the maximum output power. The narrowband source is optimized by dispersion control, and thus the FWHM is compressed to 0.20 nm under the maximum power with spectral-broadening-free property when power rising. It can be used in a spectral beam combining system to improve the diffraction beam quality.

Objective In modern frontier physics application fields, such as ultra-high resolution spectroscopy, precision measurement, laser cooling and the trapping of atoms, and optical frequency standard, very high laser performance (excellent frequency stability, narrow output linewidth, etc.) is required. However, the monochromaticity and the linewidth of the free-running laser cannot meet the application requirements, so it is necessary to stabilize the frequency and power. A common method with good maturity and a high signal-to-noise ratio is the laser frequency stabilization technology based on saturated absorption spectrum and wavelength modulation frequency locking. Therefore, it can stably lock the laser frequency for a long time. However, the laser is a type of electro-optic converter that is very sensitive to the environment. The temperature and vibration significantly influence its frequency stability, which requires good environmental adaptability by the frequency stabilization scheme that is used to improve the laser performance. As a result, this paper focuses on the environmental adaptability of semiconductor laser lock-in systems. The frequency locking scheme is based on the saturated absorption spectrum modulation and demodulation technology. The findings may be useful in guiding the stability and the practical optical path design of semiconductor laser frequency locking.

Methods In the laboratory environment, we use three typical frequency locking schemes: doppler-free locking (DFL), cell-reflection locking (CRL), and mirror-reflection locking (MRL) optical path to lock the frequency of the semiconductor laser. The three schemes use a lock-in regulator (LIR) for optical amplification. Then, through the modulation and demodulation technology to lock the laser on the saturation absorption peak. Meanwhile, the acousto-optic frequency-shifting optical path generates the laser frequency required for the experiment. Finally, we tested the environmental adaptability of the three schemes. The research method is to change some environmental parameters such as temperature, vibration frequency, and vibration amplitude to test the stability of the laser frequency and describe the environment tolerance of each scheme. A heater is used to heat the shell of the semiconductor laser in the temperature experiment, and the error signal and the temperature of the locking point are monitored. The vibration experiment is conducted on the optical platform. The influences of vibration amplitude (the driving power of vibration source) and vibration frequency on the three frequency locking schemes are studied. Because the optical platform can effectively block the high-frequency signal, only the low-frequency signal of 0--300 Hz is studied in the experiment. The duration of each vibration is set to 30 s.

Results and Discussions The temperature experiment shows that the laser unlocking temperature of the DFL scheme is 33.2 ℃, locking time is 214 s, and frequency drift is approximately 2% (Fig.4). In CRL, the unlocking temperature is 27.1 ℃, locking time is 80 s, and frequency drift is approximately 6% (Fig.5). In MRL, the unlocking temperature is 28.7 ℃, locking time is 89 s, and frequency drift is approximately 10% (Fig.6). This is because although the increase of the laser temperature leads to the fluctuation of the laser frequency and power, in DFL, because of the existence of the reference light and difference between the signals collected by two photodiodes, these interference signals can be greatly eliminated. Finally, the interference signal into the proportional integral derivative(PID) is weak. As a result, the feedback circuit can lock the laser frequency more precisely at the peak of the saturation absorption spectrum. However, as the temperature rises and reaches the critical mark, the temperature-induced interference exceeds the regulation ability of the feedback circuit. The negative feedback is destroyed, resulting in laser unlocking. Because there is no reference light for the CRL and MRL, the interference signal caused due to temperature change will enter the PID directly, making its temperature tolerance less than DFL. In the vibration amplitude experiment, the vibration frequency is set to 100 Hz. The fluctuation of the DFL optical path is relatively stable at approximately 1%, when the driving power of the vibration source increases from -4 dBm to 4 dBm; CRL fluctuates the least (approximately 0.3%), and MRL changes a lot (approximately 10%) (Fig. 7). In the vibration frequency experiment, the driving power of the vibration source is set at 0 dBm. The fluctuation of the DFL optical path is relatively stable at approximately 0.7%, when the vibration frequency increases from 0 Hz to 300 Hz; CRL has the smallest fluctuation (approximately 0.3%). MRL varies from 0.3% to 0.9% (Fig. 8). The results show that when the vibration frequency and amplitude change, the influence on DFL and CRL optical path is small and the influence on MRL is obvious. For DFL, the influence of noise caused due to vibration is greatly weakened due to the existence of reference light. For CRL, the pumping light comes from the weak reflection of the atomic gas cell, which is very weak. The fluctuation of the error signal is small due to the small saturated absorption signal and noise level, but the stability is weaker than that of DFL. The saturation absorption signal is relatively strong for MRL, and the probing light is reflected from the mirror. The small change in mirror position caused due to vibration impacts the spectral signal, increasing the noise and decreasing the stability.

Conclusions Three typical laser frequency locking schemes are studied using the experimental idea of controlling variables based on the principle of the saturated absorption spectrum and wavelength modulation frequency locking. The results show that DFL has the best temperature tolerance and antivibration interference, followed by CRL and MRL, indicating that DFL is more suitable for harsh environments. However, CRL and MRL also have certain application values. For example, because the probing and pumping lights in DFL need to coincide as much as possible, the distance between the two mirrors in the optical path must be as far as possible, and the actual optical path occupies a larger space. Therefore, DFL may not be applicable when the optical path size is limited. However, CRL and MRL have small optical path sizes and are easy to integrate and miniaturize for future laser units.

Objective Chaos synchronisation key distribution driven by a common noise light has higher security because the synchronisation coefficient between the driven signal and the response signal is low, and the eavesdropper cannot restore the complete drive signal and extract the key information from the drive signal, which is essential. Distributed feedback (DFB) laser, a vertical cavity surface-emitting laser and Fabry-Perot laser chaos key distribution systems driven by noise light have all been reported. To realise the chaos key distribution in the above-mentioned scheme, the synchronisation of the chaos and external modulator must modulate the phase polarisation or intensity state of chaos, which increases the complexity of the system, Meanwhile, it is not conducive to the construction of the key distribution system. The distributed Bragg reflector (DBR) laser is a semiconductor laser with a tunable wavelength. The laser output wavelength can be changed by adjusting the current loaded in the DBR region, and then the wavelength input in the chaotic secret distribution system of the DBR laser can be realised. Therefore, it is significance to study the chaotic dynamic characteristics of noise light injected into DBR lasers. In this paper, the chaotic dynamic characteristics of the DBR laser under the injection of noise light are experimentally studied. We found that when the noise light has a large frequency detuning with the main mode, the DBR laser exhibits a completely different chaotic state and its energy is mainly concentrated in the low-frequency range. This paper provides a foundation for high-speed key distribution technology using the synchronisation of noise-light injection common-drive DBR lasers.

Methods The noise light generated by the super-luminescent diode (SLD) is filtered, amplified and refiltered by a tunable filter (TF₁), erbium-doped fibre amplifier (EDFA₁) and TF₂ before the injection of noise light in DBR laser. The optical path uses a variable optical attenuator and polarisation controller to adjust the optical power and polarisation of the injected light. The chaotic signal of the DBR laser is amplified and filtered by EDFA₂ and TF₃, respectively. It is divided into three paths of detection by two optical couplers. In the experiment, the DBR laser gain region current is set to 57.6 mA (1.33 I_th), the DBR region current is 12.1 mA, the free-running centre wavelength of the laser output is 1549.2540 nm (frequency v_DBR) and the optical power (P_DBR) is 1.85 mw. Furthermore, the bias current of SLD is 350.0 mA, while the total power of the emitted light is 12.42 mW. Both TF₁ and TF₂ have a 3-dB filter linewidth of 6 GHz and the same centre wavelength with λ_SLD (frequency v_SLD). Additionally, the filter line width and centre wavelength of the tunable filter TF₃ are adjusted according to the state of the chaotic laser emitted by the DBR laser to ensure that the entire chaotic signal can be filtered out and the interference of the detection signal on the injected noise light (Fig. 1).

Results and Discussions The optical spectra; electrical spectra and temporal waveforms of the injected light show the injected light is a noise signal (Fig. 2). We fix noise-light injection strength κ_j=PSLD/PDBR=0.32, where P_SLD is the noise-light power entering the DBR laser. When Δv=v_DBR-v_SLD is 0 GHz, that is, under the main mode noise-light injection, the laser output is like those produced by light injection into ordinary DFB lasers. Further increase of Δv when the centre wavelength of the injected noise light is in the valley between the two modes of the DBR laser, their corresponding Δv are 27.5 GHz and 81.5 GHz, respectively. At this time, the chaotic electrical spectra are inverted “check” shape, the relaxation oscillation peak are obvious, and their 80% energy bandwidth are 1.36 GHz and 1.29 GHz. The temporal waveforms show an obvious chaotic oscillation state, but the amplitude fluctuation is small. When the centre wavelength of the injected noise light is at the peak of the DBR optical spectrum side mode, the Δv values are 47.5 GHz and 104.5 GHz. The chaotic laser electrical spectra are flat in the low-frequency band, and the oscillation peaks are almost disappeared, and their 80% energy bandwidth are 1.10 GHz and 1.07 GHz (Fig. 3). Furthermore, when the noise light is at the main mode, the chaotic bandwidth decreases as the injection strength decreases. When noise light is at the side mode, the chaotic bandwidth decreases as the injection strength increases (Fig. 4). Moreover, the correlation dimension of chaos reflects the complexity of the system. When the noise light is injected into the main mode, the correlation dimension of the chaos generated by the DBR laser is 6.17±2.02; and when the side mode is injected, it is 1.86±0.36. Meanwhile, when the noise light is in the main mode, the injection strength has little effect on the fluctuation of the correlation dimension, which shows that the chaotic complexity is higher, and the fluctuation comes from chaos itself and has little relation with external noise (Fig. 5). In key distribution system driven by noise light, the driven signal will inevitably be introduced into the detection end. This process will reduce the synchronisation coefficient of the legitimate user, and the phenomenon is not conducive to the chaos key distribution. In the case of side mode injection, the filter can avoid the introduction of noise light. Additionally, the chaos correlation dimension and the fluctuation of the dimension are small. It is easy to obtain a higher synchronisation coefficient, which is conducive to chaos synchronisation key distribution.

Conclusions Facing the application requirements of chaos key distribution, this paper experimentally studies the chaotic dynamics characteristics of noise light injected into DBR lasers. The study found that when noise light is in the main mode injection interval, the DBR laser shows a chaotic state similar to that of the light-injected DFB laser. When the noise light is in the side mode injection interval, the DBR laser shows a different chaotic state. It is mainly concentrated in the low-frequency band, and the relaxation oscillation peak is not obvious. We further studied the bandwidth and correlation dimension of chaotic lasers under different frequency detuning and injection strength and found that underside mode injection the correlation dimension is 1.86±0.36, the noise component introduced in the detection process is low, which is more conducive to the construction of chaos key distribution system.

Objective Yttrium oxide (Y₂O₃) transparent ceramics have the advantages of high melting point, good chemical stability, wide optical transparency (230 nm-8.0 μm), high infrared transmittance, low phonon energy, and high thermal conductivity. They have great application value in high-temperature infrared windows, domes, infrared detectors, luminescent media, lasers, and semiconductor industries. During the preparation process, due to the limitation of the processing technology, tiny pores may be formed on and in the grain boundaries of Y₂O₃ transparent ceramics, which causes the ceramics cannot achieve complete density. The pores and impurities significantly reduce the optical transmittance of Y₂O₃ transparent ceramics and produce large absorption, reducing the ceramics’ mechanical and thermal properties, leading to their breakage and failure in extreme environments with high temperatures, high speed, and strong impact. Therefore, it is important to measure and characterize defects, such as pore impurities, in Y₂O₃ transparent ceramics. Although, methods such as an optical microscopy, scanning electron microscopy, electronic analytical balance, X-ray tomography, and ultrasonic testing can observe the morphology of pores and other defects from the macro or micro levels, they cannot measure the absorption characteristics of defects or detect defects that are not visible in visual imaging but have abnormal absorption. When examining faults in the body, several approaches may cause sample damage. Thermal lens technology based on the photothermal effect is frequently used to identify absorption properties and defects in weakly absorbing solid materials, such as thin films and optical glass. The materials will exhibit thermal deformation on the surface or body when they are excited by a powerful pump light. The thermal lens technology can measure thermal deformation as a result of light absorption. This method has a high detection sensitivity, can precisely evaluate defect absorption properties, and offers a noncontact and nondestructive assessment.

Methods Build surface absorption and in-body absorption measurement devices based on the principle of photothermal and thermal lenses. After the pump laser is modulated by the chopper and focused by the lens, it is incident perpendicular to the surface or the body of the sample after being modulated by the chopper and focused by the lens. The temperature field of the material at the focal spot changes, causing local refractive index variations to produce a “heat slope”. After the beam is extended, the lens focuses the probe light obliquely into the sample surface or body, overlapping the focal position of the pump laser. First, the cerium oxide polishing liquid is used to polish the transparent surface of Y₂O₃ transparent ceramics on both sides and an X-ray fluorescence spectrometer is used to measure the main components. Thereafter, a scanning electron microscope is used to measure the ceramic surface morphology, an optical profiler is used to measure the surface roughness, and a spectrophotometer is used to measure transmittance. Then, the built photothermal measurement device is used to measure the stability, contrast, and absorbance scan results.

Results and Discussions It is verified that the absorption stability of the photothermal measurement system on the surface and in the body is less than 5% using fused silica glass. Comparing the photothermal measurement results with the same area of Y₂O₃ transparent ceramics using an optical microscope, the unevenness can reach 80%, indicating that the photothermal measurement system can characterize sample defects. According to the statistical characteristics, the average value plus three times the standard deviation (E+3σ) is used as the segmentation threshold. The images are binarized to indicate the location of the defect (Fig. 6). The scanning measurement of absorption in the different areas of the Y₂O₃ transparent ceramic surface shows that the absorption of different sample areas have large differences, a high degree of unevenness, and defects such as scratches (Fig. 7). Y₂O₃ transparent ceramic body absorption measurement results show that there are only small-sized pores and impurities in the body, and the proportion of defects is about 3% (Fig. 8). As a result, using photothermal scanning imaging technology, a link between the absorption signal and ceramic defect may be created, allowing for high-sensitivity detection of the defect and the evaluation of the unevenness of the sample absorption.

Conclusions The experimental results show that the absorption amplitude of surface and internal defects is significantly higher than the intrinsic absorption amplitude of ceramics, and the absorption unevenness caused by surface and internal defects is basically above 50%. The statistical distribution of absorption amplitudes on the surface and in the body shows that the low intrinsic absorption distribution is nearly Gaussian. The image is segmented using E+3σ as the threshold to determine the defect distribution area according to the statistical characteristics. According to the calculation of the binarized image, the defect area accounts for about 3%. Due to the influence of processing, the absorption amplitude and the proportion of defect area on the ceramic surface are higher than the area without processing influence. The experimental results establish the relationship between absorption and defects and realize the precise positioning of surface and internal defects, which is of great significance in improving the ceramic preparation process.

Objective The laser distance measurement system is widely used in various fields such as long-distance measurement, atmospheric concentration monitoring, three-dimensional laser imaging, and star-to-ground laser ranging, due to its small volume, high accuracy, and fast response speed. The echo time of the time-of-flight method is the focus of this technology. The situation of “strong background noise and weak signal light intensity” is that the laser ranging technology needs to face. On one hand, the background noise is too strong relative to the signal light and the signal-to-noise ratio is low, which makes signal processing difficult. On the other hand, the avalanche photon diode (APD) unit with Geiger mode is prone to channel saturation or channel blockage under strong background noise. This paper derives and discusses in detail the spatial three-dimensional convolutional neural network signal processing model, which combines the photon counting method with the convolutional neural network method. The simulation verifies that this algorithm can effectively identify the weak signal light from the strong background light.

Methods A theoretical model is established via theoretical derivation and subsequently it is verified via simulation. Firstly, different from the previous laser ranging signal processing that only considers the probability distribution of the signal response in time domain and presents the Poisson distribution, this article adds a discussion of the probability distribution in spatial domain on this basis. We believes that the signal light and background light received by the receiving lens barrel used for laser ranging have spatial distribution differences. Therefore, in the simulation, the signal light is set as the far-field spot that obeys the two-dimensional Gaussian-like distribution, and the background noise is set to be evenly distributed in spatial domain. Then, this article simulates the insertion of short signal light into long background noise, and obtains the original signal used in this algorithm with the Geiger mode 4×4 APD detection. Second, this algorithm is designed to process the spatial three-dimensional convolutional neural network signal that combines the photon number counting method in the one-dimensional time domain with the convolutional neural network in the two-dimensional spatial domain. Finally, by comparing multiple sets of simulated signals and drawing conclusions through the recognition of simulated signals, it is verified that the proposed algorithm has an improved effect on the recognition of signal light.

Results and Discussions The simulation results show that it is theoretically feasible to process the Geiger pattern array APD signal through the spatial convolution method, and the signal recognition can be nearly doubled. In the article, Figs. 10 and 11 respectively show the fitted signal obtained by simple operation and that after using 2D convolution treatment. The comparison shows that this algorithm can hide the noise in the noise. The array fitting signal is effectively strengthened to highlight the spikes and effectively improve the signal recognition. Figure 12 shows the signal diagrams fitted by using four 3×3 feature convolution kernels. Obviously, after the same 4×4 array signal is processed with different N-dimensional convolution kernels, the signal recognition is somewhat different. But the peak point is close. Figure 13 and 14 respectively show the fitted signal diagrams and signal recognition degree curves after using different N×N feature convolution kernels. The results show that as the dimensionality of the convolution kernel increases, the signal recognition first increases significantly. The big follow-up tends to be flat. Obviously, when the dimensionality of the detection signal and that of the signal processing convolution kernel matches, the signal recognition can reach the best. After using the Geiger pattern array detection device, the introduction of spatial domain signal processing method is beneficial to signal processing.

Conclusions In this paper, a three-level signal recognition optimization algorithm based on photon counting and convolutional neural network is designed by combining the existing single-photon detector array with its auxiliary circuit used in laser ranging and based on the idea of multi-level improvement of signal recognition. This algorithm not only inherits some of the calculation methods of the traditional time-of-flight algorithm, but also combines the spatial statistical distribution characteristics of the photons scattered on the plane of the two-dimensional detection array, and comprehensively considers the traditional one-dimensional signal processing method and the two-dimensional convolutional neural network. The new idea constitutes a spatial three-dimensional convolutional neural network program for processing laser ranging signals. After simulation verification, the program is feasible in theory as well as engineering.

Objective With the development of vacuum technology, subject to directional flow and uneven temperature, the thermodynamic equilibrium has been destroyed. In this case, the pressure standard has become unsuitable for characterizing the vacuum state. To improve the accuracy of vacuum measurement and the stability of measurement system, vacuum metrology will be characterized by gas density. The precisive measurement of gas refractive index based on a Fabry-Perot cavity can be used to derive the gas density. This type of vacuum metrology based on the optical method can transform vacuum metrology from physical standards based on the mercury pressure gauge to quantum standard. In recent years, quantum vacuum metrology technology based on the Fabry-Perot cavity has been widely studied; however, the laser resonance frequency at a high vacuum reference point is generally used as the reference frequency to measure the refractive index of a gas. Therefore, the measurement period of the system will be longer in the high vacuum reference point, and the cavity length will change due to gas pressure and gas penetration, which has a considerable effect on the accuracy of refractive index measurement. In this study, we report a method to measure the refractive index of gas, i.e., using a Fabry-Perot resonator for measuring the refractive index of a gas at a low vacuum constant pressure reference point. This method can shorten the measurement period and improve the measurement accuracy. We hope that our basic strategy and findings will aid in reducing the effect of cavity length change caused by gas pressure and gas permeation on gas refractive index measurement.

Methods Firstly, based on the first principles, we calculate the theoretical refractive index values at constant pressure and temperature through ab initio calculation. Then, the refractive index measurement expression of the constant pressure reference point is derived by varying the laser longitudinal mode frequency before and after inflation. In the next step, at 10 ^-5 Pa high vacuum reference point and 10 ³ Pa constant pressure reference point, using the quantum vacuum measurement device based on Fabry-Perot cavity, the laser frequency change in the cavity is accurately measured using PDH(Pound-Drever-Hall) frequency locking technology and beat frequency technology, respectively, and the helium refractive index is obtained.

Results and Discussions The refractive index measurement method of low vacuum constant pressure reference point was used for measurement, and the refractive index measurement value of constant pressure reference point was found to be closer to the theoretical value of the refractive index compared with the high vacuum reference point at the pressure level of 10 ² Pa and 10 ³ Pa. At the same pressure point, the deviation between the measured and theoretical values at the constant pressure reference point is smaller than that at the high vacuum reference point (Fig. 2 and Fig. 3) because the change in cavity length caused by gas pressure and penetration is effectively shielded, the reduction of helium in the cavity caused by helium penetration is reduced, and the measurement period is shortened in the pressure range from high vacuum to the constant pressure reference point using the theoretical refractive index of the constant pressure reference point as the refractive index measurement coefficient. In general, this method can improve the accuracy of gas refractive index measurement. In addition, the contribution of related parameters to the uncertainty of refractive index measurement is analyzed. The relative uncertainty of the theoretical value of the refractive index at the constant pressure reference point is 6.74×10 ^-14, and the measurement uncertainty of refractive index parameters is 1.28×10 ^-8 and 1.22×10 ^-8 when the cavity pressure is lower or higher than the constant pressure reference point, respectively.

Conclusions In researching quantum vacuum metrology standards, this paper proposes a method to measure the refractive index of gas and obtains the refractive index measurement expression of the constant pressure reference point. Using 10 ³ Pa constant pressure as the reference pressure to measure the refractive index, the measurement error caused by the cavity length change in the pressure range from high vacuum to constant pressure reference point is effectively shielded. Consequently, the uncertainty of refractive index measurement caused by the decrease of helium present in the cavity caused by helium penetration is reduced. The refractive index measurement of the constant pressure reference point is compared with that of a high vacuum. We found that this method can effectively improve the measurement accuracy. During the actual measurement process, we discovered that this method could shorten the measurement cycle and reduce the effect of system thermal noise. Our research shows that using appropriate pressure reference points to reduce the effect of cavity length change caused by gas pressure and gas penetration can improve the accuracy of refractive index measurement.

Objective Chemical reaction energy is converted to optical energy using a high-energy chemical laser. Its laser beam has good collimation and a high-power density when it is exported. The classic laser’s exporting window is normally made of crystalline material, but the crystalline window’s corrupt practice gradually emerges as laser power increases. There is no one type of crystalline window for the middle-infrared band laser that can withstand temperature distortion without exploding due to bulk absorption. Consequently, the free-vortex aerodynamic window (ADW), which seals the optical antrum using aerodynamics, has been commonly used. When the ADW works, ultrasonic airflow can produce an air curtain to seal the optical antrum. Simultaneously, the quality of the output beam would be affected by the gaseous aberration medium formed as the “window.” Thus, conducting a study concerning ADW’s optical quality is necessary for further improving ADW’s performance.

Interferometry, far-field method, shear interferometry, and the Shack-Hartmann (S-H) model method, among others, are available. The reference wavefront interferometry established during ADW’s gauging requires an ideal environment; the far filed process only provides the macroscopic property and cannot quantify the wavefront aberration, putting ADW’s optimization design at a disadvantage; the obtained interferometric fringe shear interference is the result of wavefront difference, in which interpreting both fringes and wavefronts is difficult. The clear aperture of the currently proposed ADW is 280 mm×10 mm, which is a large rectangle. Therefore, the S-H model approach, which uses the Zernike polynomial to rebuild the wavefront, is not appropriate. The unscanned ADW has a large length-width ratio in the paper, resulting in considerable environmental noise. An ADW optical quality detection method that can provide a reference for ADW’s engineering application is required for measuring big length-width ratio ADW’s optical quality in such a complex environment, quantitative analysis wavefront aberration; it should also provide a reference for ADW’s engineering application and have the potential to aid in the future optimization of ADW.

Methods According to the past ADW optical quality measurement and engineering application experience, the preliminary knowledge of ADW’s aberration component is already available. On this basis, an S-H splicing method is investigated in this study, which uses autocollimation S-H to measure wavefront and splicing method to rebuild wavefront; 671-nm optical source is used to verify furthermore. The experiment discusses and analyzes the peak-to-valley (PV) and root-mean-square (RMS) values in restructured wavefront when ADW is not in use and ADW’s working status is stable. This can explain the feasibility of S-H splicing method to measure ADW’s optical quality and its great significance to ADW’s optimization and engineering applications. The method also provides a new perspective to discuss big length-width ratio spot’s optical quality. The autocollimation S-H includes a light source, beam splitter prism, a battery of lenses, beam zoom implements, S-H wavefront sensor, CCD camera, and standard plane mirror; the current designed ADW’s clear aperture is 280 mm×10 mm, the pressure ratio is 100, and the working gas is N₂. The 671-nm light source goes through a battery of lenses, beam splitter prism, and beam zoom to expand a 300-mm diameter annular facula. The facula’s optical axis is parallel to ADW’s optical thoroughfare, and the facula would return the way it came after the incident the standard plane mirror vertically, which is the autocollimation process. ADW imposes restrictions on facula’s size to 280 mm×10 mm; therefore, slit facula returns the way it goes through the microlens array to focus and then image on CCD after shrinking. The autocollimation S-H wavefront sensor was adopted in this study; its microlens’ quantity is 24×24, corresponding to 300-mm diameter annular facula before shrinking, every 10 mm occupy 0.8 subaperture. Therefore, the 280 mm×10 mm rectangle facula focuses on a subspot list after passing through the microlens. To obtain more subspot, avoid the beam zoom implements second mirror block’s influence, the paper bias uses the S-H. The paper used a CCD camera’s collecting frame frequency of 120 Hz, a 2-s working duration of ADW, and subaperture’s quantity of 22. The subspot’s period is coincident and stable after wavefront going through the microlens without aberration. After ADW work, wavefront suggests aberration because of the gas medium’s supersonic flowing; each subspot appears offset with it. The splicing method rebuilds each subwavefront according to the offset between the actual spot center with reference spot center, splices each subwavefront according to the wavefront’s continuity, and rebuilds the whole wavefront.

Results and Discussions According to the measured result of the past ADW and engineering application experience, ADW’s impact on optical quality is considerably reflected in tilt, defocus, and astigmatism aberration. This study uses a 671-nm light source; bias uses a one-dimensional autocollimation S-H wavefront sensor to measure ADW and form a list of spots on CCD (Fig.1). The study aims at this list of spots and proposes the splicing method to rebuild the wavefront; simultaneously, it calculates the tilt aberration. This method covers the shortage that the S-H model method is unsuitable for silt facula (Fig.3). The contrastive paper analyzes the rebuilt wavefront’s long exposure aberration when ADW is not in use, and its pressure is normal, PV value changes from 0.0212λ to 0.1729λ, and RMS value changes from 0.0074λ to 0.0578λ (Fig.6, Table 1). The experiment data contribute to ADW’s further optimization and have great directive significance to ADW’s engineering application.

Conclusions The paper deals with the rebuilt wavefront’s long exposure when ADW is not in use and it has stable pressure. The former y tilt amount is 0.021 μrad, the PV value with a tilt is 0.0297λ, the PV value without tilt is 0.0212λ, and RMS value without tilt is 0.0074λ. The latter y tilt amount is 0.3184 μrad, the PV value with a tilt is 0.2708λ, and RMS value without tilt is 0.0578λ. The gray value of the long exposure image shows the ADW’s current working status. The paper trigger CCD to save images 1 s ahead of ADW launch so that long exposure time is 1 s when ADW is not in use. According to tilt aberration, ADW’s stably working time is 1.3 s, so that long exposure time is 1.3 s when ADW’s pressure reaches the set value and remains stable. When ADW’s pressure reaches the set value and remains stable, the rebuilt wavefront’s long exposure explains that in ADW’s 0--10 cm area, the aberration results are in a relatively large airflow. It is the direction of further optimization. While the whole PV value is controlled in less than half wavelength, it explains that the ADW meets engineering application requirements.

Objective By projecting structured light to object surface, the feature points for binocular stereo matching could be easily extracted for realizing three-dimensional (3D) measurement. Accurate extraction of the feature points is critical to the precision of 3D measurement. The encoded structured light pattern is usually used to guarantee the extraction accuracy of the feature points. In this case, the encoding modes always require the special design of the structured light pattern. And an adaptive decoding method should be implemented. However, the grid structured light has obvious intersection features which means the encoding and decoding processes are not essential. Thus, the extraction of the feature points could be simplified. In this paper, we proposed a binocular 3D measurement method using the intersections of the grid light pattern as the feature points for stereo matching. The image processing algorithm for extracting the intersections was presented. Then the stereo matching could be easily accomplished by sorting the feature points according to the grid topology. The displacement measurement experiments were carried out to verify the robustness and accuracy of the proposed method. We hope that our work could provide a binocular 3D measurement method with the advantages of high precision, strong robustness, easy to deploy, and low cost.

Methods A novel feature point extraction algorithm for grid-structured patterns was developed to extract the intersections of the grid-structured patterns in the left and right views. The algorithm implemented a coarse-to-fine process. Firstly, the regions of the feature points were locked by the improved corner extraction method. After that, the fine pixel coordinates of the feature points could be obtained by calculating the intersections of the grid lines crossed in the regions. Furthermore, the maximum value of the measurable depth of the system was analyzed. According to the working characteristics of the system, a suitable method to determine the topological relationship of the feature points was presented even while some points missed in left and right views due to occlusion.

Results and Discussions For the region location of the feature points, the corner point extraction was performed on the whole image at first. Considering the presence of noise, the common Harris corner point extraction algorithm could fail (Fig. 7). By contrast, the Shi-Tomasi algorithm could produce better results (Fig. 8). However, it is inevitable that false extraction or missing corner may occur due to the image noise (Fig. 8). Then, the density clustering algorithm was applied to solve this problem. The noise simulation experiments proved that the algorithm was capable of eliminating the false extracted points and grouping the corner points accurately (Fig. 9). After that, the extracted corner points grouped by density clustering could determine the region of each feature point [Fig. 10(b) and Fig. 10(f)]. The light stripe center extraction of the horizontal and vertical lines was performed respectively in this region [Fig. 10(c) and Fig. 10(g)]. The equations of the cross lines were fitted through the center points. In this case, the pixel coordinates of each target feature point could be calculated [Fig. 10(d) and Fig. 10(h)]. Furthermore, some feature points may miss due to occlusion which brings difficulty in stereo matching. The idea of region growth was introduced to find the relative sequence relationship between the feature points, which can effectively avoid the problem of sequential coding caused by the absence of feature points (Fig. 16). To verify the robustness and accuracy of the proposed method, the displacement measurement experiments were carried out. The measurement results of the proposed method were compared with those of the grating ruler with the accuracy of 1 μm. The maximum relative error is 2.20% (Fig. 21).

Conclusions A binocular 3D measurement method using grid structured light was proposed in this paper. The algorithm and effective stereo matching method of the feature points were studied. First, the extraction of the grid corner points was implemented by using the Shi-Tomasi algorithm. Then, the density clustering algorithm was applied to eliminate the false extracted points and group the corner points accurately. After that, each group of corner points defined a region. The light stripe center extraction of the horizontal and vertical lines was performed respectively in the region. The equations of the cross lines were fitted through the center points. In this case, the pixel coordinates of each target feature point could be calculated. Furthermore, a suitable method for determining the topological relationship of the feature points was studied even while some points may miss in left and right views due to occlusion. The verification experiment results of the proposed method were compared with those of the grating ruler with the accuracy of 1 μm. The maximum relative error is 2.20%. And 3D shape measurements of the sheet metal parts with different deformation were implemented using the proposed method.

Objective Goos and H?nchen discovered in 1947 that when an incident beam of finite size undergoes total internal reflection on the interface of two media, the actual reflection point shifts laterally along the incident plane relative to the incident point, and the shift is known as the Goos-H?nchen (G-H) shift. A finite-width incident beam can be compared to a series of plane waves travelling in different directions. These plane waves have different reflections at the interface of the two media. After the superposition of the differences in intensity and phase of all reflected light, the incident beam shifts to a certain extent in the transverse direction. G-H shift has great potential applications in optical isolation, optical sensing, and integrated optics. However, in general, the G-H shift at the material interface is very small, which is only a few times the wavelength. Therefore, it is not conducive to observation, measurement, and practical application. Hyperbolic metamaterials (HMMs) are a type of highly anisotropic uniaxial material named after their hyperbolic dispersion relations. HMMs have a wide range of applications, including light field localization, enhanced spontaneous emission, and subwavelength imaging. In this paper, we present the enhancement, direction transformation, and critical wavelength modulation of the G-H shift on the surface of the subwavelength HMM slab.

Methods Researchers proposed models including steady-state phase, energy transfer, and plane wave linear expansion functions to calculate the G-H shift of reflected light. In the laboratory, position-sensitive detectors, weak measurement, and interferometry are usually used to observe the properties of G-H shift. For nonmagnetic media, the dielectric coefficient of HMM is in the form of a second-order tensor. When the vertical component of the dielectric coefficient is positive and the parallel component is negative, its dispersion surface is a hyperboloid of bilobate type, which is called Ⅰ-type HMM. When the vertical component of a dielectric coefficient is negative and the parallel component is positive, the dispersion surface is a hyperboloid of univalent type, which is called Ⅱ-type HMM. Currently, the most common artificial HMMs include a multilayer structure of metal and dielectric stacked with subwavelength thickness and a metal nanowire array embedded in a dielectric. The multilayer HMM design is determined by the target spectral range, loss, and impedance matching. We investigate the effects of incident wavelength, filling factor, and background permittivities on the properties of G-H shift using the effective medium theory and the stationary-phase method.

Results and Discussions First, we calculated the real and imaginary parts of the equivalent permittivity of the material slab with the incident wavelength (Fig. 2) and determine the types of HMM materials at different incident wavelengths. Then, the variation of the G-H shift of different types of the subwavelength HMM slab with incident angle is calculated (Fig. 3). Figs. 3(a)--(d) present the bulk material, Ⅰ-type HMM, elliptical HMM, and Ⅱ-type HMM, respectively. Fig. 3 shows that the G-H shift of the HMM slab is 100 times that of bulk material under the same incident parameters. In addition, under the same structural parameters, the G-H shift of Ⅰ-type HMM is more significant than that of Ⅱ-type and elliptical HMM slabs.

Second, we calculated the G-H shift with the increasing incident angle when the incident wavelengths are 325 and 350 nm, respectively [Fig. 4(a)], which shows that the G-H shift in the opposite direction can be obtained only by changing the incident wavelength. Fig. 4(b) shows the variation of the reflection phase of the surface of the HMM slab with the incident angle. The phase mutation point corresponds to the G-H shift peak, and when the phase suddenly decreases, the G-H shift is positive, and otherwise,it is negative. Moreover, the relationship between the maximum G-H shift and incident wavelength is calculated (Fig. 5). We discovered that a critical wavelength exists between the positive and negative G-H shifts. The G-H shift is positive when the incident wavelength is less than the critical wavelength, and otherwise, it is negative. The greater the G-H shift is, the closer the incident wavelength is to the critical wavelength, indicating symmetry.

Finally, we studied the effects of the silver filling factor and background dielectric constant on the G-H shift characteristics (Fig. 6). We found that increasing the filling factor and the dielectric constant of the background medium is equivalent to the blue shift of the critical wavelength. When the incident wavelength is less than the critical wavelength, the direction of the G-H shift will change. When the incident wavelength is greater than the critical wavelength, the direction of the G-H shift remain unchanged.

Conclusions In the present study, the intensity and direction characteristics of G-H shift of the subwavelength HMM slab have been revealed. It shows that except enhancing the value of the G-H shift, the HMM slab can achieve the direct transformation of the G-H shift under different incident wavelengths. A critical wavelength exists between the positive and negative G-H shifts. The G-H shift is positive (negative) when the incident wavelength is less (greater) than the critical wavelength. At the same time, the closer the incident wavelength is to the critical wavelength, the larger the G-H shift is. We also discovered that the filling factor and background permittivities can be used to tune the critical wavelength. The critical wavelength presents blue-shift as the filling factor or background permittivities increases (decreases) (red-shifts). We believe that the G-H shift at the surface of the subwavelength HMM slab is very promising for potential applications in novel all-optical isolators, optical sensing, and integrated optoelectronic devices considering these intriguing discoveries.

Objective In laser inertial confinement fusion (ICF) research, the suprathermal electrons generated by the high-power laser and plasma interaction (LPI) have always been a research hotspot. Whether using a direct or indirect drive central ignition scheme, suprathermal electrons with energies greater than 50 keV will deposit some of their energy in the fuel, resulting in preheating and a reduction in implosion performance. It is discovered in the shock ignition scheme that suprathermal electrons with energies less than 100 keV generated by the spike pulse can help enhance the intensity of the shock wave and increase the energy gain. The suprathermal electrons are generally considered to be accelerated by the electron plasma waves, which are generated by LPI instability, such as stimulated Raman scattering (SRS), two plasmon decay (TPD), and resonance absorption. However, it is still unknown which instability dominates the generation of hot electrons. As a result, experimental research on suprathermal electrons’ spatial energy spectrum distribution, obtaining the main sources of suprathermal electrons, and exploring ways to control suprathermal electron generation is critical for ICF research.

Methods The experiment was carried out at the ShenGuang II high-power laser facility. The schematic of the experimental setup is shown in Fig.1. The four north laser beams (5th, 6th, 7th, and 8th beams) were used and irradiated on the 10-μm thick CH foil at 45° and P polarization direction. The wavelength of the laser is 351 nm, the pulse is a 1-ns square wave, and the energy of one beam is 250 J. The diameters of laser focal spots with or without continuous phase plate are approximately 250 μm and 120 μm, respectively. One laser beam’s corresponding average power density is 5.1×10 ¹⁴ and 2.2×10 ¹⁵ W/cm ². In the experiment, the plasma density distribution was measured with a Normaski interferometry system, and the probe beam was an 80-ps duration pulse with a wavelength of 527 nm. The suprathermal electrons were measured by two calibrated electron spectrometers (ESM A and ESM B). The spatial distribution of suprathermal electrons emitted from the target’s back was measured by a big size Image Plate, which was covered by a 50-μm thick Al foil. The backward-scattered light produced by LPI instabilities was collected by two optical fibers connected to an Ocean FX light spectrometer.

Results and Discussions The energy spectra of suprathermal electrons measured in the experiment were fitted well with the Maxwellian distribution with the electron temperatures of approximately 30--65 keV (Fig.2). It demonstrates that in all laser conditions used in our experiment, the fitted electron temperatures and intensities obtained in the normal front direction of the targets are greater than those obtained in the normal back direction. Furthermore, the angular distributions of total suprathermal electron energy in the backward direction are obtained, which have a Gaussian distribution with a peak along with the target’s normal (Fig.3). In the obtained scattering light spectra, there is a strong convective SRS signal, a TPD signal with the characteristic of a double peak (Fig.4). By analyzing the intensities of scattering light and suprathermal electron, we find that the hot electrons generated by SRS are more dominant than TPD under the condition of high-temperature large-scale plasma (Fig.6). In addition, the relationship between the kinetic energy of accelerated electrons and the wavelength of scattering light is investigated. It is found that the accelerated electron kinetic energy increases as the scattering light wavelength increases, which is consistent with the experimental observation (Fig.7).

Conclusions In this paper, we measured the energy spectrum and spatial distribution of the suprathermal electrons and the scattering light for the C₈H₈ foil targets irradiated with three laser conditions. By comparing the scattered light spectra and the hot electron kinetic energy spectra in the range of 20--500 keV, we found that the changes in the SRS scattered light intensity and amount of hot electrons with different laser conditions are synchronous, from which we speculated that SRS is more responsible for the diagnosed hot electrons than the TPD. Furthermore, the calculation confirms the hypothesis that the accelerated electron kinetic energy increases with the wavelength of the scattering light, which is consistent with the experimental observation. This conclusion implies that the development of laser-driven ICF will result in higher temperature and larger-scale plasma conditions than current experiment conditions and an increase in the contribution of SRS to suprathermal electrons compared to TPD. As a result, effective measures to control the growth of SRS, and reduce the preheating effect caused by suprathermal electrons are required.

Objective Recently, the study of integrated photonic devices is an important field of investigation, especially the photonic chip loaded with an atomic medium is one of the research hotspots. Based on the atomic medium, the integration of optical non-reciprocity, optical storage, and atomic clocks can be achieved. This requires an efficient interaction between light and atoms, which means that a certain optical depth is necessary and it is difficult to achieve at room temperature. The optical depth of an atomic medium can be increased by heating, but it is limited to the size of the photonic chip, which is not suitable in this scheme. Therefore, the use of light-induced atomic desorption (LIAD) is necessary, which does not require a heating system and it is beneficial to the integration of photonic chips.

LIAD is an impressive phenomenon observed in a sodium-vapor glass cell, whose inner surface is coated with a thin siloxane film. It has been observed in experiments that an incoherent light illuminates a cell filled with alkali atoms and these atoms adsorbed on the inner surface fall off. Thus, LIAD can effectively increase the atomic concentration in the vapor cell, which is completely different from the thermal effect, and many parameters influence the effect of light-induced atomic desorption, such as the intensity and frequency of desorption light, the atom species, the geometry of the cell, and the inner surface morphology. In this paper, it is found that the atomic concentration increases due to the LIAD effect and photothermal effect. The influences of two mechanisms are experimentally studied, which is of great significance to the research of photonic chips.

Methods A 150 mm diameter sphere with a highly reflective coating on the inner surface is used to enhance the LIAD effect, and a 460 nm light-emitting diode (LED) array mounted on the sphere is used as the illumination source. The power of the desorption light is changed by adjusting the driving current of the LED. When LIAD is carried out in a rubidium vapor cell, it is found that the temperature of the cell rises, which means that the atomic concentration increases due to the LIAD effect and the photothermal effect. By measuring the photothermal temperatures under different desorption light powers, the increase of atomic concentration caused by the photothermal effect is distinguished, and thus the relative influences of pure LIAD and photothermal effect are obtained. In order to verify whether the heat conduction caused by the LED array affects the cell or not, a water cooler is used to cool the LED array in the experiment. Besides, a glass plate is used to isolate the heat convection, which can reduce the effect of heat radiation on the cell.

Results and Discussions With the increase of desorption light power, the effects of LIAD and photothermal effect are both increased (Fig. 4). When the intensity of desorption light is 15 mW/cm², the contribution of pure LIAD is close to 6 times that of the photothermal effect. In this case, the increase in atomic concentration caused by photothermal effect is basically negligible, and when the intensity reaches 80 mW/cm², the ratio of atomic concentration increase caused by pure LIAD to that caused by photothermal effect is close to 6:4, which means that the effects of these two mechanisms are comparable (Fig. 5).

By cooling the LED array with a water cooler and isolating the heat convection, the temperature of the rubidium cell is 1--2 ℃ higher than that without water cooling (Fig. 6), which means that the heat radiation caused by the LED array has little effect on the cell. Besides, the effect of heat convection is even negligible, which means that the increase of temperature in LIAD is mainly caused by the photothermal effect.

Conclusions LIAD is a useful method to increase alkali atomic concentration without heating. In this paper, it is found that when using desorption light with high power density, the combined action of pure LIAD and photothermal effect further enhances the increase in atomic concentration. Besides, when increasing the wavelength of desorption light, the LIAD effect is reduced, while the photothermal effect is undoubtedly increased. Therefore, by changing the wavelength of desorption light, the contribution of these two mechanisms can be adjusted. In addition, since the desorption light with different power densities corresponds to different temperatures, the temperature of the atom cell can be controlled by adjusting the power density of the desorption light. These results have important reference value for the research of integrated photonic chips.

This paper develops a low-phase noise microwave frequency synthesizer for a cold atom gravimeter. The phase-locked loop technique improves the near-end phase noise at the output frequency of the 100 MHz crystal oscillator. The 6.834 GHz microwave signal required for the ground states transition of the ⁸⁷Rb atom is obtained by building an ultralow-phase-noise multiplier. The measured absolute phase noise performance of the 6.834 GHz signal is -60 dBc/Hz and -120 dBc/Hz at offset frequencies of 1 Hz and 10 kHz, respectively. The frequency resolution is as small as 1.42×10 ^-6 Hz. Meanwhile, we investigated the impact of the microwave source’s phase noise on the cold atomic gravimeter’s measurement resolution. The microwave frequency synthesizer is compact and easily expandable to other quantum precision measurement fields such as atomic clocks and atomic interferometers.

Objective Microwaves are commonly used in quantum precision measurements, and their frequency, power and phase must be precisely controlled. Atomic clocks, for example, are typically closed-loop locked using frequency hopping or phase tuning. Other experiments, such as measuring electric fields with Rydberg atoms, necessitate driving and adjusting the microwave antenna’s power. For example, in the cold atomic gravimeter, the doppler shift caused by gravity during the free fall of atoms needs to be compensated by scanning the frequency of the Raman laser and the resolution of the frequency tuning will affect the measurement precision. On the other hand, the performance of the microwave source could affect the measurement accuracy of the experiment and the stability of the atomic clock due to the Dick effect. Because of its ultra-low-phase noise performance, a photo-generated microwave has been widely used in quantum precision measurement. However, due to its complex structure, it is difficult to integrate. This paper develops a miniaturized vehicle-mounted frequency synthesizer with low-phase noise and ultra-high frequency resolution.

Methods Based on the phase-locked loop (PLL) technique, the phase noise of the 100 MHz ultra-low-phase noise crystal oscillator is improved. [Fig. 1(a)]. Next, the 100 MHz signal enters the comb spectrum generator to obtain the 6.7 GHz frequency after filtering. Finally, the 6.7 GHz signal is mixed with the direct digital synthesizer (DDS) to obtain the required 6.834 GHz for the experiment [Fig. 1(b)]. Besides, we developed a control system to tune the DDS (Fig. 2), microwave switch and attenuator to manipulate the output frequency and power of the frequency synthesizer.

Results and Discussions We adjusted the bandwidth of the phase lock loops(PLL) to 31 Hz. After PLL, the absolute phase noise of the 100 MHz signals has been improved to -96 dBc/Hz at the offset frequency of 1 Hz (Fig. 3). Mixing the output frequencies of channel 1 and channel 2 [Fig. 4(a)] to test the frequency sweep. The test results show that the signal phase remains continuous during the sweeping process [Fig. 4(b)]. The absolute phase noise performances of the 6.834 GHz are shown in Fig. 5. At offset frequencies of 1 Hz and 10 kHz, the phase noise is -60 dBc/Hz and -120 dBc/Hz, respectively. The measured results show that microwave frequency synthesis can meet the atom gravimeter’s experimental requirements.

Conclusions This paper builds a miniaturized, transportable microwave source with low-phase noise and ultra-high frequency resolution for a cold atomic gravimeter. The 100 MHz crystal is locked to a 10 MHz crystal by controlling the PLL bandwidth to improve its near-end phase noise from -73 dBc/Hz to -96 dBc/Hz at the offset frequency of 1 Hz. The reference frequency of AD9852 is overclocked to 400 MHz, to extend its frequency output range up to 160 MHz. The achieved frequency resolution is as small as 1.42×10 ^-6 Hz. The developed control system can programme the frequency and power of the microwave source to meet the needs of most quantum precision measurement experiments. The evaluation results show that the microwave frequency synthesizer can satisfy the cold atomic gravimeter’s requirements of a μGal level measurement precision. Furthermore, the PLL and frequency multiplier scheme can be easily extended to other target frequencies without loss of generality.

Objective An imaging spectrometer is the fusion of imaging technology and spectral technology that uses multiple channels to detect targets. We can obtain spatial information and spectral information on the target at the same time. According to the different spectral characteristics of different objects, targets can be identified and analyzed in detail to obtain more comprehensive data. An infrared dual-band imaging spectrometer can detect and recognize targets with high precision and resolution in a complex environment. It also has the advantages of high accuracy and low false alarm rate. A middle wave infrared imaging spectrometer is used to detect the spectrum of radiation from high-temperature objects. It is mainly used to detect volcanic activity and red alert for forest fires in civil use. It can be used in the military field to detect high-temperature exhaust gas from aircraft and tanks. A long wave infrared imaging spectrometer is used to detect objects at normal temperatures. It is very useful in mineral resource exploration and atmospheric gas detection. It is also used in the military to identify camouflaged targets with strong stray radiations. Therefore, research on infrared dual-band spectral imaging technology is of great significance to the development of military and civilian fields.

Methods The front telescope objective system, the spectral spectroscopic system, and the secondary imaging relay system of the infrared dual-band common image plane imaging spectrometer were designed using the modular design method. Because the infrared dual-band system had wide spectra and chromatic aberration was difficult to correct, the front telephoto objective adopted an off-axis two-mirror system with few degrees of freedom. The spectroscopic system adopted an Offner convex grating structure with less smile. We designed the diffractive order of the system. The diffractive order of the middle wave band was second order, and the diffractive order of the long wave was first order. The dual diffraction orders could obtain a better spectral resolution, and make more effective use of the detector in the meantime. The blazed wavelength of the convex grating was the center wavelength of the middle wave band, which could ensure a high diffraction efficiency for both bands. To reduce the influence on the imaging results of stray light reaching the detector, we designed the secondary imaging relay system to ensure that the optical system showed 100% cold-stop efficiency. Each part of the system had telecentricity. Through pupil connection and matching, the overall system design was completed. The imaging quality of the system was evaluated and analyzed through the spot diagram, modulation transfer function (MTF) curve, and distortion curve. We analyzed the narcissus phenomenon of the system. Through real ray tracing, the values of YNI and I/IBAR were obtained for each surface of the transmission system. If one of the parameters is greater than 1, no obvious narcissus will occur. If the values of the two parameters are both less than 1, it is necessary to analyze the narcissus phenomenon further through reverse tracing.

Results and Discussions The finally designed infrared dual-band common image plane imaging spectrometer has good imaging quality. The spectral resolution of the imaging spectrometer is high, the spectral resolution of the center wavelength of the middle wavelength band is better than 5 nm, and the spectral resolution of the center wavelength of the long-wavelength band is better than 10 nm (Fig. 16). We design the blaze wavelength of the convex grating to ensure the dual-band diffraction efficiency of the imaging spectrometer is relatively high and the diffraction efficiency of the middle wave band and the long wave band are both higher than 80% (Fig. 10). The narcissus phenomenon of the refrigerated imaging spectrometer was analyzed and the results showed that the system had no obvious narcissus phenomenon (Fig. 18).

Conclusions In this paper, an infrared dual-band common image plane imaging spectrometer was designed for the refrigerated dual-color quantum-well infrared detector. The pixel size of the detector is 25 μm, the array size is 384 pixel×288 pixel , and the working spectral range includes middle wave 4.4--5.4 μm and long wave 7.8--9.2 μm, F nuber is 2.5. Through the modular design method, the front telephoto objective system, Offner spectroscopy system, and relay system are designed separately. The front telephoto objective system is an off-axis two-mirror system with a spatial resolution of 0.1 mrad. The dual-band diffraction efficiency of the Offner spectroscopy system is higher than 80%, and the smile and keystone are small. The relay system achieves 100% cold-stop efficiency. The three systems compose the overall design of the refrigerated infrared dual-band imaging spectrometer through pupil connection and matching. The design results show that, in the full wavelength band and full field of view, the root mean square radius of the spot diagram is smaller than the size of a single pixel of the detector, the modulation transfer function is close to the diffraction limit, the spectral resolution is better than 10 nm, the distortion is less than 1%, and there is no obvious narcissus phenomenon. The imaging quality of the system is good, which meets the design requirements of the infrared detection system.

Objective N, P, and K, the three main nutrient elements of compound fertilizer products, are essential nutrient elements for crop growth; therefore, the quality inspection of compound fertilizers is very important. Currently, the main methods for detection of compound fertilizer include flame atomic absorption spectrometry, inductively coupled plasma emission spectroscopy (ICP-AES), and near infrared reflectance spectroscopy (NIRS). The detection time of these methods is short. However, samples must be pretreated, thereby preventing real-time monitoring of the samples, and possibly reducing the accuracy of the measurements. Compared with the above technologies, laser-induced breakdown spectroscopy (LIBS) technology requires no sample pretreatment and is characterized by a green and safe detection process, short detection time, and real-time detection of all elements. This technology has been used in various fields. When LIBS technology is combined with partial least squares to model and predict the nutrient elements of compound fertilizers, numerous samples are usually needed to improve the prediction accuracy of the model. The accuracy of prediction is generally low when the sample size is small. Therefore, improving the prediction accuracy for small sample sizes is important. In this article, we propose a data extraction method based on statistical principles to expand sample spectral data for small sample sizes, thereby improving the measurement accuracy.

Methods Twenty types of compound fertilizers, with N, P₂O₅, and K₂O as the main components, were investigated in this work. A LIBS detection system was set up for collecting and analyzing the radiation spectrum of the plasma; then, we used a new proposed method of data extraction. Afterward, the N, P, and K elements in the compound fertilizer samples were modeled and predicted using a partial least square method combined with principal component analysis. The last 75 sets of spectra were used as T2 spectra to obtain the relative error between the true N, P, and K element content of each sample and the predicted content. To determine the robustness of the calibration model, we randomly removed five samples for modeling and predicting the N, P, and K content of these samples.

Results and Discussions The spectral data were preprocessed. Compared with the original spectrum, the background of the preprocessed spectrum is eliminated, and the relative intensity between channels changes, but the relative intensity of each spectral line between the channels remains unchanged (Fig. 3). The N, P, and K elements are modeled and predicted after the preprocessing steps. The coefficients of determination for N, P, and K element content modeling in the training set are 0.99, 0.98, and 0.99, respectively, and the root mean square errors are 0.4309, 0.0979, and 0.3385, respectively; moreover, the coefficients of determination obtained for the fitting curves of the predicted and true values of the N, P, and K element contents in the prediction set are 0.99, 0.98, and 0.99, respectively, and the root mean square errors are 0.4787, 0.0706, and 0.0195, respectively (Fig. 5, Fig. 7, and Fig. 9). The average relative errors between the true and predicted values of the N, P, and K element contents of 20 samples obtained from T2 spectra are 2.33%, 0.70%, and 3%, respectively (Fig. 6, Fig. 8, and Fig. 10). After the sample data are expanded, the average predicted relative error (ARE) values of N, P, and K elements in the 20 compound fertilizer samples are all <3%, and ARE of P is <1%. Compared with the unexpanded condition, the average relative error of the measured element content dropped by more than 10% (Table 2). The robustness of the model is determined. The predicted values of N and P content agree well with the true values, the relative errors are all below 12%, and the relative errors of K element are mostly above 20% (Table 3, Table 4, and Table 5).

Conclusions In this work, the partial least square quantitative analysis method is used to establish a regression model, and a data extraction method based on statistical principles is used to expand the small sample size of the compound fertilizer spectral data. The N, P, and K element content of a compound fertilizer sample is modeled and predicted. The average relative errors of the content prediction in 20 samples are 2.33%, 0.70%, and 3.00%, respectively. Further, the robustness of the model is determined by randomly removing the data of five samples. The results reveal that the predicted values of N and P element content concur with the actual values, and the relative errors are mainly below 12%. Thus, after using the data extraction method based on statistical principles to expand the sample spectrum data, the average relative error of the measured element content is reduced by more than 10% compared with the unexpanded time. The experimental results show that when the sample size is small, the accuracy of the measurement can be improved using this new data extraction method combined with the partial least square quantitative analysis model for regression modeling.

Objective Guanine is one of the four bases of deoxyribonucleic acid (DNA). It pairs with cytosine in the double helix structure of DNA to maintain the stability of life activities. However, guanine methylation can affect the normal operation of DNA. When guanine is methylated, it immediately depurinates and forms apurinic sites, causing DNA alkylation damage and increasing cytotoxicity. One of the byproducts of guanine methylation, 7-methylguanine (7-MG), is commonly used as a biomarker to assess alkylation damage. However, traditional medical methods, such as gas chromatography-mass spectrometry (GC-MS) and high-performance liquid chromatography (HPLC) used to detect 7-MG, are time-consuming, cumbersome, and costly. Therefore, medical research needs a new accurate and swift method to detect guanine methylation. Furthermore, THz fingerprint spectral characteristics enable it to effectively identify biomolecules. However, the detection limit of the traditional tablet pressing method is at milligram level, which cannot meet the application requirements of low concentration detection (microgram and less) in the biomedical field. Some researchers have proposed combining terahertz spectroscopy and metamaterial biosensors; however, these metamaterial biosensors are limited to the detection of a pure substance and cannot realize qualitative identification of substances and mixed quantitative analysis. The chip designed in this study was tested on binary and multicomponent mixtures to check if it could predict the concentration of 7-MG in mixture samples. Finally, the 7-MG content of the mixture was determined using the standard internal method and the variation function of a pure 7-MG product. The minimum detection limit is 6.30 μg, which is 500 times lower than 2.95 mg by the traditional tablet pressing method. Furthermore, when 7-methylguanine and other substances are mixed together, they exhibit different frequency shift changes on the chip, allowing high sensitivity qualitative differentiation and quantitative detection from the mixture. This study provides important reference value for the subsequent rapid detection of 7-MG content in human cell DNA, and the detection and treatment of diseases.

Methods In this paper, 7-methylguanine is considered as an example to design a metamaterial chip based on the capacitance and inductance effect to enhance THz detection sensitivity. First, the frequency shift response of 7-methylguanine and guanine was measured through terahertz time-domain spectroscopy. The chip used in this study was then tested on binary mixtures and multicomponent mixtures to check if it could predict the concentration of 7-MG in mixture samples. Finally, the 7-MG content of the mixture was determined using the standard internal method and the variation function of a pure 7-MG product.

Results and Discussions In this paper, the relation between chip frequency shift and the amount of 7-methylguanine (Fig. 4) is obtained by conducting experiments. When the amount of sample increases, i.e. When the concentration increases, the characteristic peak starts moving to a lower frequency (red shift). The nonlinear equation f(x)=aexp(bx)+cexp(dx) is obtained through function fitting, where f(x) is the frequency shift and x is the effective mass of the sample. The corresponding coefficients are: a=0.04336, b=0.002242, c=-0.04559, d=-0.05696, and the goodness of fit of determination coefficient R²=0.9915 can be obtained. The same test is performed on guanine (Fig. 5), the corresponding coefficient of guanine was a=0.06861, b=0.002499, c=-0.06831, d=-0.02969, and the goodness of fit of determination coefficient R²=0.9895 was obtained. These results show that for known samples, the content can be detected using the corresponding frequency shift relation; for unknown samples, the curve of unknown samples can be deduced by testing samples of different concentrations and fitting the frequency shift curve and comparing with the existing curve parameters to achieve qualitative analysis of the unknown sample. From the test of the binary mixture, the frequency shift of the mixture is found to be the superposition of the frequency shifts of individual substances in each group (Fig. 6). Then, the same conclusion was made by the testing the multicomponent mixture (Fig. 7). This shows that the frequency shift effect of each component in the mixture can be separately calculated, and the frequency shift amount follows the frequency shift rule of pure product. Simultaneously, by comparing the actual value and the calculated result of the mixture frequency shift, clearly, the total frequency shift of each substance is almost the superposition of the single-frequency shift of each substance, with an accuracy >85%.

Conclusions This paper provides a new method for nondestructive, rapid, and accurate detection of molecular methylation. Considering 7-MG and G as examples, a terahertz metamaterial chip is designed based on capacitive and inductive effect. The detection limit of the chip can reach 6.30 μg, which is about 500 times smaller than that of 2.95 mg measured using the traditional pressing method. Here, the metamaterial chip is covered with different concentrations of 7-MG and G. The specific change in absorption peak frequency shift allows for qualitative and quantitative analysis of 7-MG and G. Furthermore, the mixture test confirms that the frequency shift of the mixture is a superposition of the frequency shift of a single substance. Then, using the standard internal method and the variation function of a pure 7-MG product, the content of 7-MG in the mixture can be calculated. This method can also be used to identify other molecular methylation products, such as 6-methylguanine, which is formed through guanine methylation, and 5-methylcytosine, which is formed through cytosine methylation. Hence, the findings of this study can be used in the future to accurately detect human DNA methylation.

Objective The quantum coherent effect of electromagnetic induced transparency (EIT) is characterized by a narrow transmission peak in a broad absorption band. This phenomenon is associated with the slow light effect, which can be used in optical buffering, refractive index sensing, and other applications. And the development of the EIT effect has been limited due to the extremely difficult implementation conditions. Metamaterials are artificial composite materials made from natural materials with unique physical and chemical properties that natural materials do not have. Adjusting the resonant frequency of bright and dark modes to make them resonant at a close frequency and then combining them to produce an atom-like electromagnetic induced transparency phenomenon is how EIT-like behavior in metamaterial is generally realized. However, in some reported research, the passive modulation method is used in EIT-like metamaterial, which greatly limits the application of related devices. Therefore, studying EIT-like metamaterials in the terahertz band under active regulation is an important research topic.

Electronic control is now widely used due to its simple operating conditions. In general, electronic control can be achieved by incorporating electrically adjustable devices, such as variograms and other electrically tunable materials. The reported graphene structures are typically polarization- and incident-angle-sensitive, with the transmission peak disappearing under oblique incidence. Because of numerous possibilities for the polarization direction and angle of the incident wave in practical applications, polarization- and angle-insensitive metamaterial devices are more suitable for application requirements. The proposed structure shows excellent characteristics, such as polarization-independence, incident angle-insensitivity, obvious slow light effect, and high refractive index sensitivity.

Methods A novel polarization- and angle-insensitive metamaterial structure based on the cross and 4L-shaped graphene is designed. CST Microwave Studio software is used to run all numerical simulations. By splitting the structure and analyzing the surface current and electric field distributions at different frequencies, the physical mechanism is discussed. The effect of geometric parameters, such as cross and 4L-shaped graphene lengths on a transparent window are investigated. Finally, the characteristics of the proposed structure, such as polarization- and angle-insensitivity, tunability, slow light effect, and refractive index sensitivity are studied.

Results and Discussions The structure, which consists of a cross and four L-shaped graphene resonant units, has an obvious EIT-like effect, with a peak value of more than 0.8 (Fig. 2). The resonant dips of the structure with a single cross or 4L-shaped graphene are observed dips near 1.81 THz and 2.19 THz, respectively. The transparent peak lies between the resonance frequencies of the two isolated resonators (Fig. 2). At the frequency of 1.75 THz, strong electric fields are concentrated at both ends of the horizontal cross, and the surface current moves unidirectionally on the cross graphene (Fig. 3). Therefore, the cross surface experiences electric dipole resonance. The surface current and electric field generated along the arms of the 4L structure are the same at the 2.18 THz resonance frequency, which is also known as dipole resonance (Fig. 3). Both cross and 4L structures can be directly excited by incident waves as bright modes exhibiting electric dipole oscillations for the transmission peak near 1.94 THz. Surface currents are excited on the cross and 4L-shaped graphene structures simultaneously, and the directions are anti-parallel (Fig. 3). As a result of the destructive interference caused by the coupling of the two bright modes, a distinct transparency window is formed.

When it comes to structure parameters, the high-frequency transmission dip has a slight redshift as the length of 4L-shaped graphene increases (Fig. 4). As the length of the cross increases, the low-frequency transmission dip in the transmission spectra is slightly red-shifted (Fig. 5). To investigate the effect of the distance between the cross structure and the 4L structure on the transparent window, transmission spectra, and electric field intensity distributions under the peak at various distances are drawn. The transmission spectra and electric field intensity distributions show very little difference at different distances. This demonstrates that the metamaterial under consideration in this paper is fault tolerant (Fig. 6). As the Fermi energy of graphene increases, so do the transfer window is blue shifted (Fig. 7). The corresponding frequency modulation depth is approximately 0.286, with the Fermi level varying between 0.3 eV and 0.6 eV. The maximum amplitude modulation depth is 0.724, with graphene has Fermi level ranging from 0.4 eV to 0.5 eV. Because the structure has symmetry, the metamaterial structure proposed in this paper is polarization insensitive (Fig. 8). When the incident angle is less than 60°, there is no significant change in the transmission spectra for both TE polarization and TM polarization. This shows that the structure is insensitive to the incident angle of the incident wave (Fig. 9). When the Fermi level of graphene is 0.5 eV, the maximum delay of the transmission peak is 0.81 ps and can be controlled by tuning the Fermi energy (Fig. 10). The designed structure has a refractive index sensitivity of 395 GHz/RIU, which is obviously higher than many conventional EIT-like structures and can also be regulated by tuning the Fermi level (Fig. 11). Thus, it has great potentials in the field of refractive index sensing applications.

Conclusions Finally, this paper investigates metamaterials made up of the cross and 4L-shaped graphene resonant units. Numerical simulations show that bright-bright mode interference produces the EIT-like effect in the terahertz band. The studied metamaterial has great polarization- and angle-insensitivity characteristics. The transmission spectrum does not change significantly when the incident angle is less than 60°, and the transmission peak can be kept above 0.75. By tuning the Fermi level of the graphene, the frequency modulation depth of 0.286 and the amplitude modulation depth of 0.724 are achieved. In addition, the proposed structure has obvious slow light effect and high refractive index sensitivity. Meanwhile, the actively tuned group delay and refractive index sensitivity are achieved by changing the Fermi level of the graphene. These characteristics can be used in many applications, such as modulators, switches, light buffers, and sensors in the terahertz band.