Silicon based integrated optical devices have the outstanding advantages of small footprint and high integration level. They have broad application prospects in the fields of optical communication, data center optical interconnection and so on. However, the size of silicon-based waveguide coupler is relatively large, and the working bandwidth and process tolerance are limited. The design of silicon-based multimode routing photonic devices still faces big challenges. In this paper, two advanced design methods are introduced for silicon-based integrated optical devices developed in recent years: shortcuts to adiabaticity method and transformation optics method, their physical principles are briefly expounded, and typical applications in the design of silicon-based integrated optical devices are presented.

With the explosive growth of network transmission data, traditional integrated circuit is facing difficulties of further increasing exchange rate and expanding chip capacity. Compared with traditional electronic chips, silicon photonic devices present the advantages of high speed, low power consumption, larger bandwidth, lower cost, good compatibility with CMOS process, thus they show great potentials to meet the requirements of next-generation all-optical switching networks, the optical interconnection for data centers and the high-performance computing. Therefore, silicon photonics is widely recognized as one of the most promising solutions to break through the bottleneck of capacity in the post-Moore era. In this work, technical principles of silicon-based optical switch and switch arrays are introduced, especially MZI and MRR switch units and the construction strategy of switch array. Progresses and proposal of promising solutions to the challenges of silicon optical switches are reviewed.

Silicon-based photonic chips use photon as the information transmission media, which has the advantages of broad bandwidth, high speed, high integration and compatibility with CMOS technology, and has application value in many fields. In this paper, first, several manufacturing platforms of silicon-based photonic chips are introduced, including silicon-on-insulator (SOI), SiN, Ⅲ-Ⅴ compound materials (such as GaAs and InP) and lithium niobate film on silicon, then the development status and challenges of silicon-based photonic chips are reviewed in the fields of optical communication and interconnection, optical computing, biosensors, light detection and ranging (LiDAR)and quantum photonics, and finally, a summary is presented.

Microwave beamformer is one of the core components in phased array radar systems and 5G communication base stations. In recent years, silicon integrated microwave photonic beamformers have become one of the research hotspots in microwave photonics due to their advantages of large bandwidth, compact size, light weight, low loss, and anti-electromagnetic interference. In this paper, the basic principles and the key performance indices of microwave photonic beamforming systems are first introduced. Then, several integrated tunable optical true delay lines and optical beamforming network architectures that applied in the integrated microwave photonic beamformers in recent years are summarized. And also, the recent progresses in the integrated microwave photonic beamforming chips and the automation control and calibration methods are reviewed. Finally, the prospects of the future developments of silicon integrated microwave photonic beamformers are prospected.

To meet the requirements of a high-gain and miniaturized receiver optical subassembly (ROSA) for radio over fiber (RoF) systems, a four-channel ROSA device with a size of 20.0mm×14.0mm×5.9mm is designed and fabricated based on hybrid integration technology. Low noise amplifier (LNA) chips are integrated into the module to improve the radio frequency (RF) signal gain, and the RF signal transmission circuit is established to simulate the performance of the device. The RF signal gain of the device is 14dB, the -3dB bandwidth of the device is 23GHz, and the responsivity of the device is 0.81A/W at the incident light wavelength of 1550nm. The crosstalk of RF signals between adjacent channels is less than -40dB. This module has important significance for reducing the volume and power consumption of the RoF system.

According to the large junction capacitance of the rib waveguide pn junction of the single-drive push-pull silicon modulator, which limits the further improvement of the modulators bandwidth, a silicon modulator with slot rib waveguide pn junction structure is proposed. By etching the rib waveguide pn junction with a certain size of slot, the capacitance of the pn junction is reduced and the modulation bandwidth is increased. The two pn junctions are simulated and compared, and the results show that with the same electrode structure and doping conditions, the electro-optical bandwidth of the slot rib waveguide modulator is increased by about 8.3GHz compared with the rib waveguide modulator. Based on the optimized single-drive push-pull slot rib waveguide modulator, the 90Gbit/s OOK signal modulation is simulated, and 14.69dB extinction ratio of eye-diagram is obtained.

New generation on-chip sensing system puts forward the requirements of miniaturization, integration, low cost, etc. Silicon-based integrated waveguide devices adapt to its development trend, and the sub-wavelength grating structure is favored in the field of refractive index sensing because of its unique mode distribution characteristics. In this paper, first, the operating principles of the sub-wavelength grating waveguide are introduced, focusing on its advantages in the field of refractive index sensing. Then, the latest research progress of refractive index sensing based on subwavelength grating structures are sorted out according to the classification of device structures, and their advantages and disadvantages are analyzed and summarized simultaneously. Finally, the future development of refractive index sensing based on sub-wavelength grating structures is discussed.

Since the breakthrough of epitaxy technology of Si-based Ge film, Si-based Ge optoelectronic devices have developed rapidly, among which Si-based Ge photodetectors are the most prominent. Because Ge can absorb the light in the near-infrared communication band and is fully compatible with the Si CMOS process, Si-based Ge detectors are almost the only choice for Si-based optical detection. This paper mainly introduces the research progresses of two common Si-based Ge photodetectors (normal incidence and waveguide coupling), including typical device structures, and the main methodes to improve the performance of responsivity and bandwidth.

In order to meet the requirements of co-packaged optics (CPO) system, a multi-quantum well (MQW) continuous wave (CW) high-power distributed feedback (DFB) laser with high power, low noise and low loss is designed with the wavelength of 1310nm. There is a specially optimized far-field reduction layer (FRL) below the MQW active layers to pull optical mode into the lower n-cladding layers. This design can reduce the internal loss of waveguide and the optical confinement factor. The fabricated DFB laser exhibits high slope efficiency and enables uncooled CW operation at high ambient temperatures. Test results indicate that the output power, threshold current and slope efficiency of the DFB laser are achieved to 173mW@400mA, 20mA and 0.46mW/mA at 25℃, respectively. When the injection current is 300mA, the far-field patterns divergence angle is 15.2°×19.1°, the sidemode suppression ratio (SMSR) is higher than 55dB, the relative intensity noise (RIN) of laser is less than -155dB/Hz, and the Lorentz linewidth is less than 600kHz. The output power and the threshold current of the DFB are achieved to 112mW@400mA and 32mA at 85℃, respectively.

Multi-channel interference (MCI) laser is a novel monolithically integrated widely-tunable semiconductor laser. It applies multi-arm interference based mode selection, and achieves wide tuning through either electro-optic effect or thermo-optic effect. It has the advantage of simple fabrication and large fabrication tolerance. The active and passive integration of the MCI laser utilizes the offset quantum well scheme, which reduces the difficulty for regrowth. The wavelength tuning range of the electrically tuned MCI laser is greater than 48nm, the output power is greater than 20mW, and the Lorentzian linewidth is less than 350kHz. The wavelength tuning range of thermally tuned MCI laser is greater than 60nm, and the Lorentzian linewidth over the entire tuning range is less than 100kHz at dual temperatures. Meanwhile, the research advances its stability control and integration, and demonstrates nano-ITLA based on thermally-tuned MCI lasers.

Proposed and designed is a high coupling efficiency apodized one-dimensional grating coupler based on X-cut lithium niobate insulator platform. The overlay amorphous silicon is employed to enhance the coupling efficiency by combing subwavelength photonic crystal design and genetic algorithm. The optimized grating coupler exhibits a peak coupling efficiency of -1.08dB, which can be further improved up to -0.53dB by introducing a bottom metal mirror, with a 1dB bandwidth of 61nm at central wavelength of 1550nm.

The general methods for reducing the noise of avalanche photodiode (APD) and improving its gain and bandwidth were analyzed. Aiming at the inherent contradict ion between the responsivity and bandwidth, the solutions were proposed in this paper, such as using three-dimensional nano-technology to reduce the noise of APD and using integrated optical technology to improve the responsivity. The research progresses of integrated high-speed APD were summarized, including the following three aspects: hybrid integration of surface plasmon focusing structure with APD of vertical light incidence, hybrid integration of super lens with APD of vertical light incidence, and monolithic integration of arrayed waveguide grating (AWG) and evanescent wave coupling waveguide APD of side optical incident. The development trends were prospected as well.

Silicon photonic waveguides, exhibiting high refractive index contrast and nano-scale optical field, are capable to producing compact and efficient nanophotonic devices. However, birefringence is inevitable for high refractive index contrast waveguides, and thus almost all nanophotonic devices are polarization sensitive. Polarizing beam splitter (PBS) is an important component to overcome the strong polarization dependence of silicon nanodevices in polarization diversity photonic integrated circuits, and has received much attention in recent years. The recently reported sophisticated subwavelength grating waveguide (SWG) based PBS has a working bandwidth of more than 200nm and an extinction ratio of more than 20dB. In this paper, the working principles of various types of PBSs are briefly deswoedcribed, and their performances in terms of size, extinction ratio and bandwidth are illustrated, and the advantages and disadvantages are also analyzed. Finally, the application scenarios and future development of PBS are summarized and prospected.

Proposed is a novel high-sensitivity sensor structure with a parallel bulk acoustic wave resonator (BAWR) and Ga2O3-based MSM UV sensor, namely the bulk acoustic wave resonance solar-blind UV sensor (BAWR-UV sensor). The MSM Ga2O3 UV sensor and BAWR sensor were designed and fabricated. The measured light and dark S parameters were packaged in the simulation software as S1P data model instead of MSM equivalent circuit, MBVD model was used instead of BAWR equivalent circuit, and the two structures were built in parallel to develop the BAWR-UV simulation model. The influence of the BAWR core parameters on the sensitivity of the BAWR-UV sensor was simulated by ADS software, and the design scheme to improve the sensitivity was studied. The simulation results show that in the frequency range of 0.5~4.5GHz, the larger the Qp of BAWR-UV, the higher the impedance sensitivity, and the impedance sensitivity increases with the decrease of frequency, and the optimal impedance sensitivity is 100kΩ/(μW/mm2) when fp=0.533GHz; the frequency sensitivity increases with the increase of frequency, and the best frequency sensitivity is 1.4MHz/(μW/mm2) when fp=4.5GHz. And also the exploration of high-sensitivity solar-blind UV sensors was carried out.

With the rapid development of all-optical networks, fast and slow light technology has gained widespread attention. In this paper, a theoretical model of assist light injection into semiconductor optical amplifier (SOA) was constructed to study the fast and slow light effect. The effects of the gain coefficient and the magnitude of DC current, modulation current, modulation frequency and phase shift on the phase delay with and without the injection of assist light were compared. The simulation results show that the output signal light changes from slow light to fast light under the injection of assist light, and the stronger the assist light power, the more obvious the effect. Meanwhile, it can be observed that after the injection of the assist light signal, the larger the DC current, the larger the phase delay. On the contrary, the larger the modulation current, modulation frequency, phase shift and gain coefficient, the smaller the phase delay.

Due to the merits of low loss, low cost, low power consumption and easy fabrication, polymer-based planar lightwave circuit (PLC) devices play important roles in optical communication networks, microwave photonics and optical sensor system. In this paper, the recent progresses of polymer-based PLC devices are reviewed, including array waveguide gratings, variable optical attenuators, optical switches, and their integrated devices. Finally, the future development trends of the polymer-based PLC is discussed.

In low temperature environment, the thermal expansion coefficient and thermal optical coefficient of fiber Bragg grating (FBG) material will change, which will affect its temperature sensing characteristics. In this paper, the temperature sensing characteristics of bare FBG sensors and brass tube-encapsulated FBG sensors were studied experimentally. The results show that in the temperature range of 80~300K, the sensitivity of the bare FBG temperature sensor is 6.43pm/K and the linearity is 0.974, and the relationship between temperature and fiber grating center wavelength is nonlinear in the temperature range of 80~230K. The brass tube-encapsulated FBG temperature sensor has a sensitivity of 26pm/K and a linearity of 0.996 in the whole temperature range, which is greatly improved compared with that of the bare FBG temperature sensor. Comparative experiments show that the encapsulating of fiber Bragg grating can not only improve its temperature sensitivity and linearity, but also improve its temperature sensing characteristics.

Based on the energy band theory and the principle of distributed Bragg mirror (DBR), the reason for the large DBR series resistance in VCSEL was analyzed. The potential barrier at the interface of heterojunction in DBR structure was reduced by component gradient, the doping concentration of each layer of DBR was optimized, and the heterojunction potential barrier in DBR was further reduced by regulating Fermi energy level, thus the series resistance of DBR was effectively reduced. Al0.22Ga0.78As/Al0.9Ga0.1As was used as two materials for DBR growth. The thickness of each layer of DBR was designed and the optimum growth temperature of AlGaAs material was studied. The 795nm VCSEL mutation DBR and gradual DBR were grown by MOCVD epitaxy. The resistance of mutant DBR and gradient DBR was 6.6 and 5.3Ω, respectively. The resistance of DBR after optimized growth was effectively reduced.

The effects of barrier layer thickness, gate width length ratio and doping concentration on the transfer characteristics and transconductance curve of AlGaN/GaN HEMTs devices were studied by using Silvaco TCAD device simulation software. The results show that the change of barrier layer thickness can adjust the current switching ratio and threshold voltage of the device, and realize the transformation from depletion device to enhancement device. The ability of the gate voltage to control the two-dimensional electron gas in the quantum well increases with the increase of the gate width length ratio. The doping of barrier layer increases the output current and the peak value of transconductance, but the doping of too high concentration makes it difficult to turn off the device. This is of practical guiding significance for the practical preparation of GaN based HEMTs devices.

In this paper, a pixel level pulse-frequency-modulation (PFM) ADC was designed for the wide-dynamic-range short infrared focal plane array (IRFPA). When the conventional PFM ADC uses DI to detect weak signal, the injection efficiency will decrease, which result in the loss of the dynamic range. In the designed PFM ADC, the CTIA is used to maintain high efficiency when detecting weak signal. Two levels of gain and self-selected gain technology are used to extend the dynamic range without increasing the power consumption. Self-selected gain technology enables each pixel to choose the gain according to the photocurrent automatically. The circuit was designed and simulated with 0.18μm CMOS process model. The dynamic range is up to 101.5dB, and the power consumption of each pixel is only 10.54μW. A 32×32 pixel level digital readout circuit with the pixel pitch of 50μm was designed based on the proposed PFM ADC.

A new type of spiral photonic crystal fiber based on Archimedes spiral is proposed, and silica is chosen as the background material. The cladding is composed of 24 spiral arms, each spiral arm contains 11 small air holes, and there is a big air hole in the center of the fiber core. The annular area between the cladding and the air-core can transmit the orbital angular momentum mode. It can support stable transmission of 22 kinds of orbital angular momentum modes over the wavelength range of 1300~1800nm. The effective refractive index difference can be up to 2.89×10-3, the lowest dispersion can reach 66.4ps/(nm·km), and the nonlinear coefficient is minimally up to 2.17W-1·km-1 at 1550nm wavelength. The dispersion variation is less than 15.15ps/(nm·km) in the wavelength range of 1500~1600nm. The spiral photonic crystal fiber is not only simple in structure, but also has low nonlinearity and flat dispersion performance, which provides an idea for the design of spiral photonic crystal fiber.

Three methods of hydroxylation of the microchannel plate substrate were studied. The water contact angle of the substrate surface of the microchannel plate after hydroxylation and the morphology change of the channel end surface were measured, and the hydrophilicity and corrosion of the microchannel plate substrate in various methods were analyzed. The experimental results show that the aqueous solution of ammonia and hydrogen peroxide does not greatly improve the hydrophilicity of the substrate surface, the NaOH solution has a corrosive effect on the substrate, and the hydrophilicity of the substrate surface treated with the piranha solution is significantly improved without corrosive effect. The effects of the immersion time and immersion temperature of the microchannel plate in piranha solution on the hydrophilicity of the surface were studied. The results show that with the increase of the immersion temperature, the water contact angle on the surface of the microchannel plate first decreases and then increases, and reaches a minimum value when the temperature is 80℃. The immersion time has little effect on the hydrophilicity of the surface of the microchannel plate. Finally, the surface hydroxylation process of the microchannel plate is determined as the immersion temperature of 80℃, and the standing time of 20~60min.

An active smoothing technology for the surface MSF error of high steepness aspheric elements was proposed. By constructing the off-axis machining contact model of aspheric element, the distribution of misalignment in the machining process was obtained, and the active deformation grinding disc was designed. The finite element analysis method was used to simulate the relationship between the material, thickness and structure of the grinding plate in the process of active smoothing, and the optimal active smoothing parameters were obtained. The experimental results verify that the accuracy of the calculation model and the active smoothing technology play a better smoothing effect on the MSF error of high steepness aspheric components.

By utilizing a developed electromechanical coupling model considering the coupled flexoelectric and strain gradient effect, the dimension-dependent physical fields such as displacement, electric potential, electric displacement and carrier distribution were systematically investigated. It is shown that the flexoelectricity and strain gradient have a significant influence on the electrical related physical fields, but less effect on the mechanical displacement. Furthermore, in contrast to the strain gradient enhancing effect, the piezoelectric properties of micro-nano structures are suppressed by flexoelectricity. This study provides a theoretical guidance for analysis of the electromechanical characteristics of piezoelectric micro-nanostructures.

A small building target detection algorithm is proposed based on FCOS neural network. Aiming at the problem of insufficient target features of small buildings extracted in the feature extraction stage of the FCOS algorithm, multi-scale convolution and deformable convolution were used to improve the ability of the network to extract features of small buildings. And through the improved SGE attention mechanism, the weight of interference noise in the feature map becomes lower. The improved network can extract more target feature information of small buildings and is more robust to noise. Experimental results obtained from the building dataset made by ourselves show that, under the same test environment, the network’s overall detection accuracy (mAP) of buildings under normal, dense and occluded conditions is improved by 1.7%, and that of small buildings is improved by 3.6%, reducing the missing and error detection of small building targets.

In order to solve the problems of insufficient scene adaptability and poor real-time performance of common automatic white balance (AWB) methods, an adaptive AWB method based on color channel histogram reconstruction is proposed, and the hardware circuit of the proposed algorithm is implemented by field programmable gate array (FPGA). It not only corrects the white balance of the image, but also ensures the high-speed real-time image processing of the system. In this method, firstly, the image is processed by restricted limited adaptive histogram equalization (CLAHE) to improve the image contrast, then the channel partition statistics of gray level interval is carried out for the image, and the reconstruction method of color histogram matching or translation is adopted for the images of different scene categories. Experimental results show that compared with AWB algorithm based on light source estimation, the accuracy of color temperature correction of the proposed algorithm is improved by 14%, which has better adaptability to different color scenes and real-time processing ability.

Presented is a method for processing polarised images based on non-uniform circularly polarised light irradiation for turbid underwater imaging environments. First, the difficulties encountered in imaging and processing targets under non-uniform circularly polarised light illumination conditions are analyzed and the characteristics of circularly polarised light as a light source are described. Secondly, an imaging environment is designed to acquire polarised and visible images of underwater targets. A method based on CLAHE and MSRCR is proposed for pre-processing visible light images to substantially eliminate the effect of non-uniform illumination on image sharpness. The method is shown to be able to fuse the polarisation information image and enhance its edge contours, and finally to fuse the polarisation information fused image with the optimised visible light image, taking into account the visual characteristics of human eyes. The results show that the method achieves the enhancement of the edge contours and textures of the polarised images, and significantly improves the quality of the underwater images.

In order to solve the problem of imaging noise of multi-channel time delay integrated charge coupled device (TDICCD) splicing remote sensing camera, and improve the signal-to-noise ratio (SNR) and imaging quality, two main reasons for the noise of multi-TDICCD imaging were analyzed. First, the unsynchronized operation of the CCD causes the ground plane high-frequency disturbance of the power supply of the TDICCD imaging circuit, which affects the acquisition of the effective video signal of the adjacent channel CCD. Secondly, the switching noise disturbance of DC-DC power supply in the circuit affects the acquisition of CCD effective video signal. Based on the practice of engineering development, the multi-TDICCD imaging circuit system is improved by taking measures such as sharing a unified system clock and the same system reset between CCD channels to suppress the imaging crosstalk between channels. The interference of power supply noise on multi CCD imaging is suppressed by filtering the main interference source of DC-DC power supply and multi-point grounding of TDICCD port. The imaging and signal-to-noise ratio of the improved multi-channel TDICCD imaging circuit system were tested. The results show that the measures remove the imaging noise effectively and significantly improve the SNR by 2dB, and the excellent field imaging quality can meet the requirements of practical engineering.

Compared with the traditional imaging technology, the light field imaging technology can utilize the propagation direction information of the light in the light field, and the computational imaging method can greatly improve the depth of field of the traditional imaging system. The higher the resolution of the traditional optical microscope, the smaller the depth of field. Therefore, in this paper, the light field imaging technology and traditional optical microscope was combined by inserting a microlens array on the primary image plane of the microscope, to improve the depth of field of the microscope, and realize the three-dimensional measurement of light field microscopy. The system can obtain the four-dimensional light field information of the light field through a single exposure, complete the light field microscopic measurement through digital refocusing technology and sharpness evaluation function. The experimental results show that the light field microscopy measurement method based on the microlens array is feasible. The measurement system increases the depth of field of the microscope head by nearly 100 times at the expense of 16 times the lateral resolution.

In response to the diverse applications of avalanche photo-diode (APD), a design solution of bias circuit module with multi-mode control was proposed. Based on the working principles of APD, the correlation among gain coefficient, reverse bias voltage and operating temperature was analyzed. The bias circuit modules with multi-mode control were designed, the software design was realized on the single chip microcomputer, and the module was analyzed and verified by experiments. Experimental results show that the bias circuit module with multi-mode control and temperature compensation is available in the temperature range of -40~+60℃, the deviation of the output voltage is less than 0.5V, and the ripple is less than 100mV. The module can realize APD multi-mode operation and meet engineering applications.