It is important to study the miniaturization of sun sensors used in the micro-nano satellites whose load requirement is rigorous. In order to solve the integration compatibility problem of optical devices and processing circuits, a novel on-chip sun sensor based on standard CMOS process is proposed in this paper. The micro-wall structure was constructed by metal layers, and the pn junctions were uniformly distributed on both sides as photoelectric sensors. The angle of incident light was calculated by detecting the ratio of photocurrents on both sides. The rationality and feasibility of the device were verified by the process realization, model establishment, numerical simulation and experimental testing in this paper. Finally, the on-chip sun sensor array chip is designed and fabricated whose mass is 1.5g, size is 304.2mm3, the detection accuracy is ±1.6° and the field of view range is 80°, which can meet the requirements of miniaturization.

On-chip integrated digitization is one of the main development directions of the infrared focal plane array, and its key technology is the integration of analog-to-digital conversion ADC blocks inside the readout circuit. In response to the digitization requirements of the line linear InGaAs focal plane, a non-binary redundant bit SAR ADC design was adopted in this paper. The entire readout circuit included a readout circuit unit and an analog-to-digital conversion unit. The readout circuit unit adopted the structure of CTIA, which had good structural linearity and high injection efficiency. The analog-to-digital conversion unit used a SAR ADC, which had simple structure and low power consumption. In this paper, the CDAC module was designed using a non-binary calibration method, which inserted redundant bits in the capacitor array to increase the conversion speed and accuracy of the ADC. And the sampling accuracy was improved using the bottom plate sampling technique. A 14-bit SAR ADC was designed and simulated with 0.18μm CMOS process model. At a sampling rate of 1MS/s, the SNR is 74dB and the ENOB is 13.4bits.

A package solution based on fourth-order pulse amplitude modulation (PAM4) optical transmitting components is proposed, whose data transmission rate can reach 200Gbit/s, to meet the design requirements of 200Gbit/s PAM4 optical transceiver modules. The package solution integrated four PAM4 electrical/optical conversion channels, and the data transmission rate of each single channel was 50Gbit/s. Based on the packaging solution of optical transmitting components, firstly, the components and technical difficulties of 200Gbit/s PAM4 optical transmitting components were introduced. Then, according to the technical difficulties, the 50Gbit/s data transmission channel was modeled, simulated and optimized. Finally, the test of the sample was completed. The test results show that for this solution, the data rate of a single PAM4 channel can reach 50Gbit/s, and the overall data rate of the PMA4 channels can reach 200Gbit/s, which can fully meet the design requirements of optical transceiver modules.

With the rapid development of precision manufacturing industry, the demand for precision measurement technology is increasing fast. Displacement detection technology is the basis of geometric precision measurement and is widely used in the field of contemporary precision manufacturing. The chromatic confocal displacement measurement system has the advantages of insensitivity to stray light, tilt of the measured object, and materials type, as well as high measuring frequency and high resolution. It can detect displacement, surface roughness, three-dimensional topography and the thickness of single-layer or multi-layer transparent materials. So it plays an significant role in the field of precision displacement measurement. In recent years, the chromatic confocal measurement technology using diffractive optical elements to improve the performance of optical system has been widely studied by domestic and foreign scholars. In this paper, the research progress of chromatic confocal displacement measurement based on diffraction dispersion principle was summarized. Firstly, the principle of chromatic confocal displacement measurement and the characteristics of diffractive optical elements were introduced. Then, the development history and research progress of chromatic confocal displacement measurement technology based on diffractive dispersion principle were described. Finally, the development trend of this technology was prospected.

A high dynamic range CMOS image sensor based on the time domain is proposed. The sensor was based on a novel structure that could detect high-input light intensity in the time domain and low-input light intensity in the analog domain. A time domain measurement circuit was added to the traditional CTIA structure, which could achieve a continuous and large dynamic range without changing the original integration process. Based on the 0.35μm and 5V-CMOS process, a 256×1 linear CMOS image sensor was implemented with a photodiode area of 22.5μm×22.5μm, and the photoelectric response of the device was fully characterized and simulated. The simulation results show that the time domain CMOS image sensor can reach the dynamic range of 96dB, and the two output signals of time domain and analog domain can be output synchronously with the power consumption of 7.98mW.

Frame transfer charge coupled device (FTCCD) is mainly used for visible light detection, which is prone to bloom in strong light applications. To solve this problem, a frame transfer vertical anti-blooming CCD image sensor was developed. The array size of the device was 1024×1024 elements, and the pixel size was 12μm×12μm. In order to realize the vertical anti-blooming function, an n-type buried channel saddle p-well-n-type substrate structure was adopted. The vertical anti-blooming multiple is 200X, the full capacity is no less than 200ke-, the readout noise is no more than 80e-, and the dynamic range is no less than 2000∶1.

An O-band 4-channel wavelength division (de)multiplexer with low insertion loss and flat-top transmission spectra is presented based on the novel scandium (Sc)-doped aluminum nitride (Al1-xScxN) photonics platform, and an optimization method is proposed. The designed multiplexer was based on cascaded Mach-Zehnder interferometer (MZI) filters, using directional couplers with curved waveguides to reduce wavelength dependence. Particle swarm algorithm (PSO) was adopted to increase the efficiency of device optimization according to the characteristics of coarse wavelength division multiplexing (CWDM) applications. The transmission quality of four channels was improved by adjusting the design parameters of the device structure. For Sc doping concentrations of 0%, 9% and 23%, the designed multiplexer has 1-dB bandwidth as wide as about 15.6nm and negligible insertion loss of less than 0.1dB, showing “box-like” transmission response. The crosstalks of four channels are less than -30.6 dB.

A photoelectronic reversible logic gate applied to neuromorphic photonics and quantum computation was constructed with a hybrid nano-thin film lithium niobate and silicon nitride waveguide. The main structure of the photoelectronic reversible logic gate was composed of two cascaded Mach-Zehnder modulators. The total length of the gate was only 4.4mm, being only one hundredth of the length of a common proton exchange lithium niobate modulator. Operating at the wavelength of 1.55μm, the Mach-Zehnder modulator achieved a complete power exchange with only 4.9V voltage that was well compatible with CMOS process. The device characteristics show that the photoelectronic reversible logic gate can realize reversible logic operation. Moreover, in the wavelength range of 1.4μm to 1.6μm, the mean insertion loss of the device is 0.6dB, the minimum cross-talk of the output port is -47dB, and the maximum extinction ratio is 41dB. In the voltage range of 4V to 6V, the mean insertion loss is 0.63dB, the minimum cross-talk of the output port is -26dB, and the maximum extinction ratio is 22dB. Those performances guarantee the effectiveness of the photoelectronic reversible logic gate.

An infrared absorption model for microbolometer detection units with an umbrella-shaped absorption layer structure are established. Based on the optical admittance matrix method and impedance matching theory, the infrared absorption characteristics of the model under different aperture sizes and shapes of the umbrella structure are simulated using three-dimensional electromagnetic simulation software CST. The results show that the optical performance of the dual-layer umbrella-shaped aperture microbolometer is closely related to the aperture size of the umbrella structure. The introduction of the aperture in the umbrella-shaped microbolometer reduces the heat capacity of the detection unit while maintaining high infrared absorption characteristics, thereby improving device response speed. The resulting detection unit has an average absorption rate of 85% in the 8~14μm infrared band, meeting the design requirements of large-scale small-pixel non-cooled infrared focal plane detectors.

A novel Mach-Zehnder interferometer (MZI) structure combined with multi-mode interference is proposed and applied to vibration detection. Different from the traditional MZI which was based on the cascade of two 3-dB single-mode optical coupler (OC), the MZI was composed of multi-mode OC and single-mode OC. A 3dB multimode OC with the coupling ratio of 1∶1 was fabricated using the step index multimode fiber with a core diameter of 105μm and the graded index multimode fiber with a core diameter of 62.5μm, and the MZI was constructed with the multimode OC and another 3dB single-mode OC respectively. For comparison, two single-mode OCs were also used to construct the MZI. Vibration was applied to an interference arm of the MZIs constructed with different types of OCs, and data acquisition and analysis were carried out by an oscilloscope. The results show that the MZI based on a step index multimode coupler with a core diameter of 105μm has the largest detectable vibration frequency range of 1~20kHz, and has the best performance in vibration frequency deviation and vibration frequency discrimination ability.

In order to investigate the influence of divergent ratio on the flow and heat transfer characteristics of half-corrugated microchannel, five models of half-corrugated microchannel with different divergent ratio were designed, flow (200＜Re＜1000) and heat transfer behavior were researched by numerical simulation method, and the results were compared with smooth microchannel. The results show that the pressure drop and Nusselt number of half-corrugated microchannel increase with the decrease of the divergent ratio. Compared with smooth microchannel, half-corrugated microchannel has better heat transfer performance but higher pressure drop loss. By introducing η as the comprehensive performance index, it is found that reducing the divergent ratio is an effective way to improve the comprehensive heat transfer capacity of the half-corrugated microchannel. And when the Re＞490, its comprehensive heat transfer capacity is better than that of smooth microchannel.

The finite element method was used to study the problem of tensile vibration of p-type GaN piezoelectric semiconductor rod under the action of simple harmonic force. Numerical solutions for the nonlinear dynamic responses of displacement, electric potential and hole concentration were obtained. The modulating effect of simple harmonic force on p-type GaN piezoelectric semiconductor rod was analyzed. The results show that the simple harmonic force significantly affects the distribution of physical fields in the rod. The distributions of the electric field and hole concentration lose their symmetry or anti-symmetry due to the electrically nonlinear term in the current density. The electric fields exhibit periodic variations driven by the simple harmonic force. However, the dynamic response of hole concentration exhibits significant asymmetric fluctuations.

In order to obtain the performance changes of the satellite-borne high-speed operational transimpedance amplifiers after a long period of operation in the on-board environment, experimental studies were conducted on the ionization total dose damage characteristics and change patterns of key characteristic parameters of three high-speed operational transimpedance amplifier chips with a gain bandwidth product greater than 1GHz. The irradiation tests were completed using high temperature accelerated evaluation on a 60Co γ-ray source at a dose rate of 0.3~0.5Gy(Si)/s to the amplifier chip. The characteristics of key electrical parameters such as output bias, output noise and bandwidth of the amplifier chip before and after irradiation and before and after annealing at high temperature (85℃±6℃) were analysed after the tests, and the mechanisms causing the changes in the electrical parameters were discussed. The results show that after two rounds of 150Gy(Si) dose irradiation and high temperature annealing, the output bias and output noise level of the amplifier chip do not change significantly, the time domain impulse response was normal and the -3dB bandwidth was reduced by about 3%. Bandwidth is a radiation-sensitive parameter for the three high-speed operational transimpedance amplifier chips, and its variation is directly related to the positive charge build-up and interfacial state induced by ionising radiation at the SiO2/Si interface. The irradiated chip is still capable of meeting the demands of high bandwidth test situations, with 150Gy(Si) being the cumulative dose for electrical parameters and functional qualification.

In order to achieve highly sensitive and reliable sensing detection of refractive index, a metal-insulator-metal (MIM) waveguide coupled rectangular cavity structure is proposed based on sensing performance of Fano resonance in this paper. With the help of theoretical analysis of the coupling mode and the numerical simulation by the finite element method, the Fano transmission spectral characteristics of the rectangular cavity with and without embedded metal plate are analyzed. And the parameters of the designed rectangular cavity embedded metal plate structure are optimized further. Finally, the influence of structural parameters on its sensing characteristics is discussed. The results show that two Fano mode resonant peaks are formed after light is incident into the rectangular resonant cavity with TM mode. In the first mode, the FOM is 9.4×104 and the sensitivity is 700nm/RIU; in the second mode, the FOM is 8.4×103 and the sensitivity is 1200nm/RIU. The results indicate that bimodal detection can be achieved in this structure and a high quality factor can be achieved also in each model, which provides a theoretical basis for the design of high performance micro/nano optical refractive index sensor.

A three-dimensional model of atomic layer deposition (ALD) reaction chamber was established. The influence of process parameters such as pressure, precursor pulse time and temperature on precursor distribution in ALD process was simulated and analyzed by ANSYS Fluent software. The simulation results show that the lower the reaction pressure is, the higher the diffusion coefficient of Mg(Cp)2 precursor molecule will be. And the longer the precursor pulse time is, the more even the distribution of it in the reaction chamber will be. When the pulse time is 250ms, the distribution of Mg(Cp)2 in the reaction chamber was basically even, and the mass fraction of the precursor in each part of the reaction chamber was basically the same. When the pulse time is 200ms, H2O is basically evenly distributed in the reaction chamber. In the ALD temperature window of MgO film, the diffusion effect of Mg(Cp)2 precursor molecule is stronger with the higher temperature in the reaction chamber.

In this study, based on the sol-gel spin coating method, aluminum and nitrogen co-doped TiO2 film samples were prepared with aluminum doping amount of 3.00at%, and nitrogen doping amount of 6.00at%, 7.00at%, 8.00at%, and 9.00at%, respectively. The test results of the samples show that the basic structure of TiO2 is still retained in the co-doped samples, and the grain size of aluminum and nitrogen co-doped samples is reduced to different degrees, which makes the surface of the samples modified to become more uniform and flat compared with that of the undoped. The co-doped samples showed different degrees of red shift and enhanced absorbance in both UV and visible regions. When the doping amount of nitrogen is 7.00at%, the diffraction peak of (101) is the largest, the peak type is the sharpiest, and the obtained TiO2 film has the best optical performance. Compared with the eigenstate TiO2, the band gap values of the co-doped samples are decreased, and the minimum value is 2.873eV. The above results indicate that Al/N co-doped TiO2 film can improve its optical properties.

Thin-film transistors (TFT) with top-gate and top-contact structure were prepared using zinc-tin-oxide (ZTO) film as the channel layer and polymethyl methacrylate (PMMA) film as the dielectric layer at low temperature (100℃), and the effect of oxygen partial pressure on the device performance was investigated during the ZTO channel layer formation process. The results show that ZTO channel layer has stable amorphous structure and high visible light transparency (average transmittance ≥89.61% in the range of 400~700nm), and increasing the oxygen partial pressure is beneficial to the improvement of visible light transparency. Hall test results show that the increase of oxygen partial pressure (from 3.5×10-2Pa to 7.5×10-2Pa) will reduce the ZTO electron carrier concentration (from 4.73×1015cm-3 to 6.11×1012cm-3), resulting in a reduction in the energy consumption of TFT devices based on ZTO channel layer (shown as the reduction of the off state current and the forward shift of the depletion device threshold voltage). In addition, increasing the oxygen partial pressure is also conducive to the optimization of the channel/dielectric layer interface state (that is, the subthreshold swing is reduced).

Ultra-broadband (UWB) over fiber technology has attracted much attention in recent years because of the advantages of large bandwidth and high transmission rate of UWB signals and the advantages of long-distance transmission and low loss of optical fiber transmission.A system for generating UWB signal based on microwave photonic technology was proposed, which was based on non-linear PM-IM conversion to generate monocycle signal. Then the first-order difference was performed on the signal by using a balanced photodetector (BPD) and an optical delay line to build a delay line filter, producing a doublet signal that met FCC mask requirements. The doublet UWB signal has a center frequency of 5.4GHz and a 10dB bandwidth of 4.3GHz. A signal processing technique based on DS-BPSK was also proposed, with channel coding using convolutional coding, followed by chirped Bragg gratings for dispersion compensation of the signal. Through the simulation by Opti-system software, when the FEC threshold is 3.8×10-3, the sensitivity of the receiver is increased by 5.5dBm, effectively reducing the bit error rate and improving the transmission performance of the system of ultra-wideband over fiber.

In order to improve the error-correction performance of polar codes with the medium and short code length, an improved decoding algorithm of successive cancellation list of polar codes based on error set (ES-SCL) is proposed. The algorithm first constructed an error set according to the channel characteristics of polar codes. Then the parity check (PC) bits were set according to the elements in the error set during the encoding of polar codes and the information bits and frozen bits were placed in the remaining positions. When the decoder decoded the PC bits, each path did not perform the path splitting and pruning, but could get the estimation of the PC bits through the check function. For the other bits, the decoder would directly perform the SCL decoding. Simulation results show that under the additive white Gaussian noise channel, when the code length is 512, the code rate is 0.5, the block error rate is 10-5 and the maximum list length is 8, compared with the PC-PSCL and CA-SCL decoding algorithm, the proposed ES-SCL decoding algorithm can achieve a gain of about 0.18 and 0.15dB. When the code length is 256, the code rate is 0.5, the block error rate is 10-5 and the maximum list length is 8, compared with the CA-SCL and PC-PSCL decoding algorithm, the proposed ES-SCL decoding algorithm can achieve a gain of about 0.3 and 0.35dB. Furthermore, the ES-SCL decoding algorithm that uses the partial bit splitting decoding can reduce the sorting number of about 50% when the block error rate is almost the same as that of the PC-PSCL decoding algorithm, thus its decoding complexity is lower.

In order to solve the problem that the accuracy of heart rate measurement is poor under external interference or dim environment by image Photoplethysmography (iPPG), an adaptive heart rate extraction algorithm was presented and was verified on an embedded hardware platform. Different scenes could be identified by this algorithm according to the chromaticity relationship between faces and the background areas in the image. Subsequently, the appropriate camera was launched for image acquisition and adaptive mapping. Finally, the extracted signal was filtered and the result was output after the signal quality evaluation. The algorithm was implemented on the embedded Zynq system with dual cameras for real-time heat rate measurement, and the results were visualized. Experimental results show that the measurement error decreases from 3.36BPM to 2.78BPM under the interference of light and motion, and the accuracy is improved by 17.3%. In addition, the designed system can also achieve heart rate acquisition under extreme dark conditions, with an average error of about 2.39BPM.

A 3D reconstruction model based on attention and intermediate fusion representation is proposed, aiming to reconstruct a 3D model with refined structure. This method used the axial spatial attention mechanism to learn information in different directions, and embeded it in the encoder to capture local structural features. Then, based on the two-stream network, the depth map and 3D average shape could be inferred to design an intermediate fusion representation module which could effectively fuse visible surface details to better describe the 3D spatial structure of the objects. The experimental results show that the axial spatial attention mechanism and the intermediate fusion representation module proposed in this paper enhance the ability of feature extraction. The IoU and F-score are 1.3% and 0.4% higher than PixVox++, respectively, proving that the 3D reconstruction effect is better.

Aiming at the problem of poor real-time performance due to the time-consuming implementation of the star identification algorithm for the all-sky regions in DSP, a star identification algorithm for all-sky regions based on FPGA in star radiation mode is proposed and implemented. This method took advantage of the characteristics of parallelism and high main frequency of FPGA, and solved the problem of long time-consuming of star identification in the all-sky regions under the premise of ensuring high precision and high accuracy of the algorithm. The main works of this paper were as follows. Firstly, the star radiation pattern was constructed according to the star point information. Secondly, the calculation of the corresponding eigenvectors in the star radiation pattern was realized based on FPGA. Finally, a pre-readable and pipelined star radiation pattern matching algorithm was proposed, which greatly shortened the time of star radiation pattern matching. The experimental verification was carried out on the Xc7a75tfgg484-2 platform of Xilinx Company. The results show that under the same main clock frequency, the speed of the star identification algorithm based on FPGA is more than 10.95 times faster than that based on DSP. And the recognition rate exceeds 99% when using three sets of eigenvectors for simultaneous matching.

In order to improve the segmentation accuracy of single stage instance segmentation and improve the situation of missed and wrong detection of small targets, an improved YOLACTR algorithm is proposed based on YOLACT algorithm. The algorithm first used the combination of CNN and Transformer to design a new head prediction network to further extract features, and used two-way attention to correlate the mask information of the same instance and distinguish the mask features between different instances. It paid attention to the correlation information around the feature points, making the prediction of the detection box more accurate. Then the mask branch was formed by the combination of multi-level up sampling module and the designed CS attention module, which integrated a variety of different scale information. Then the CS attention module was used to pay attention to different scale information. On the MS COCO data, compared with YOLACT algorithm, YOLACTR algorithm improves the detection accuracy of box and mask by 7.4% and 2.9% respectively, and improves the detection accuracy of small targets by 18.9% and 13.5% respectively. Experiments results show that YOLACTR algorithm can improve the accuracy of detection, segmentation and classification in multi-target complex scenes, which improves the problem of missed and wrong detection of small targets and overlapping targets.

An improved orthogonal matching pursuit (OMP) algorithm is proposed to solve signal loss problem of the fiber Bragg grating (FBG) sensing signal with external disturbances. Based on the essential characteristics of the wavelength drift of FBG sensing signals with stress, within the framework of compressive sensing theory, the effective sparsity of FBG signal was obtained by removing the imaginary part of sparsity coefficient and applying exponential saturation method fitting and ranking the non-zero elements. Moreover, by improving the atom selection strategy and the termination conditions in the iteration process of the traditional OMP algorithm, the complexity of algorithm was reduced and the reconstruction accuracy of the signal was improved. The results of the comparative experiments demonstrate that the proposed algorithm has outstanding advantages on time complexity, signal noise ratio (SNR) and reconstruction accuracy.

Due to the complex computation of the deep learning network and the huge computational parameters used in the road vehicle target detection, the problem of high delay and slow processing speed exists in the target detection task on the embedded system based on ARM architecture. Aiming at the above problems, a complete embedded road vehicle target detection solution was designed and implemented in this paper. The structural re-parameterization is used in the YOLOv3-Tiny-based feature extraction network to improve the model detection accuracy, and the parallel acceleration of the convolutional neural network was deployed by Vitis-AI on the Zynq embedded platform with the DPUCZDX8G architecture acceleration core, and finally the improved YOLOv3-Tiny network model was quantified, compiled and deployed as a dynamically linked library. The experimental results show that the MAP of VOC2007 is 0.597, and the real-time processing speed is 27.7FPS. At the same time, the frame rate power consumption ratio is 1.49, which is suitable for the low power consumption requirements of edge computing devices.

Aiming at the problems of low illumination, large change of target scale, serious occlusion between targets, difficult feature extraction of existing target detection network, poor detection effect, etc. in complex mine environment, an improved S3-YOLOv5s mine personnel protection equipment detection algorithm is proposed. A simple, parameter free attention module (SimAM) was added to the backbone network to improve the feature extraction capability of the network. Scale equalizing pyramid convolution (SEPC) was introduced to strengthen multi-scale feature fusion. Finally, SIoU was used as the frame regression loss function and K-means++ algorithm was used for prior anchor frame clustering to improve the frame detection accuracy. The experimental results show that, compared with the existing YOLOv5s algorithm, the average detection accuracy of the proposed algorithm in all categories is improved from 89.64% to 92.86%, and the algorithm has excellent detection capability for personnel protection equipment under complex mine environments, which verifies the effectiveness of the proposed method.