Abstract:We present the design， simulation， and experimental validation of an InP/In_0.53Ga_0.47As laser power converter for the wavelength of 1 550 nm. By optimizing the thickness of the absorption layer and adopting a dual-layer anti-reflective structure （SiO₂ and SiN）， the device achieved an absorptance of 96% under 1 550 nm laser irradiation， demonstrating insensitivity to angle variation and robust to wavelength shifts. The experimental results are in good agreement with the theoretical calculation results. The external quantum efficiency （EQE） reaches 92%. Under a laser power density of 47 mW/cm2， the cell’s conversion efficiency reached 23%. Further refinement of device processing is needed to reduce series and shunt resistances， thereby enhancing the overall efficiency of the laser photovoltaic cell. In addition， this study delves into the impact of cell area on the photovoltaic performance， providing optimization directions for the miniaturization of laser photovoltaic cells.

Abstract:The thermal load of the cryogenic infrared detector Dewar is a comprehensive indicator characterizing the adiabatic capacity of the Dewar. Radiation heat is a part of the thermal load. When calculating the radiation heat， the traditional approach typically simplifies the Dewar to a coaxial cylindrical model. This simplified model differs significantly from the actual one and the traditional approach is incapable of computing the radiation heat transfer between surfaces where emissivity， transmittance， and reflectance vary with wavelength. To enhance the calculation accuracy of the Dewar''s radiation heat， based on Monte Carlo principle， a 3D Studio Max model was employed， model information was extracted， and a program was developed， resulting in a set of general calculation program for the Dewar''s radiation heat based on the radiation transfer factor. To preliminarily verify the accuracy of the calculation program， the cold side radiation heat of two types of experimental Dewars was calculated according to the gray body assumption and compared with the measured values. The theoretical calculated value and measured value of the cold side radiation heat of experimental Dewar 1 were 155 mW and 136 mW， respectively， with the error of 19 mW； the theoretical calculated value and measured value of cold side radiation heat of experimental Dewar 2 were 87 mW and 79 mW， respectively， with the error of 8 mW. After initially testing the accuracy of the calculation program， the cold side radiation heat of an engineering typical 1 K×1 K long-wave Dewar was calculated and measured experimentally when the emissivity of the window facing it was 0.9. The theoretical calculated value was 127 mW， and the measured value was 110 mW， with the error of 17 mW.

Abstract:The application of InGaAs focal plane arrays （FPAs） requires high density and small pixel pitch. However， as the pixel pitch decreases， the pixel coupling becomes stronger. By fabricating 5 μm pitch InGaAs arrays with different scales， the pixel coupling effects in high-density InGaAs arrays were studied. Innovatively， matrix equations were introduced to describe the contributions of dark current from each part， and a mathematical model of pixel coupling was constructed， and the contributions of the dark current resulting from the coupling effects were quantitatively analyzed. The results indicated that at a bias voltage of -0.1 V， reverse-biased pixels in the array can reduce the dark current of adjacent reverse-biased pixels by 21.39% of the pixel''s initial dark current. In contrast， zero-biased pixels can increase the dark current of adjacent reverse-biased pixels by 219.42%. Based on this high-density focal plane pixel coupling model， the impact rules of pixel coupling on dark current have been summarized， providing new insights for dark current research in high-density InGaAs focal plane arrays.

Abstract:The performance of detectors is one of the key factors for space target detection. In this paper， the performance requirements of blocked impurity band detectors for the ground-based detection scenario of space targets are analyzed. The theoretical calculation of background radiation and point target radiation is carried out by establishing the radiation transmission model of ground-based detection scenario. The correlation between radiation and detector performance is also analyzed. Taking space debris as a typical target and ground-based telescope as a carrying platform， the key performance requirements such as quantum efficiency， dark current， full well and specific detectivity are analyzed and calculated in the mid-latitude and high-altitude detection environment. This work lays a theoretical foundation for the detector structure design of ground-based detection.

Abstract:Airborne area-array whisk-broom imaging systems typically adopt constant-speed scanning schemes. For large-inertia scanning systems， constant-speed scanning requires substantial time to complete the reversal motion， reducing the system''s adaptability to high-speed reversal scanning and decreasing scanning efficiency. This study proposes a novel sinusoidal variable-speed roll scanning strategy， which reduces abrupt changes in speed and acceleration， minimizing time loss during reversals. Based on the forward image motion compensation strategy in the pitch direction， we establish a line-of-sight （LOS） position calculation model with vertical flight path correction （VFPC）， ensuring that the central LOS of the scanned image remains stable on the same horizontal line， facilitating accurate image stitching in whisk-broom imaging. Through theoretical analysis and simulation experiments， the proposed method improves the scanning efficiency by approximately 18.6% at a 90° whisk-broom imaging angle under the same speed height ratio conditions. The new VFPC method enables wide-field， high-resolution imaging， achieving single-line LOS horizontal stability with an accuracy of better than 0.4 mrad. The research is of great significance to promote the further development of airborne area-array whisk-broom imaging technology toward wider fields of view， higher speed height ratios， and greater scanning efficiency.

Abstract:Self-supervised pre-training methods have strong capabilities in feature extraction and model transfer. However， current pre-training methods in multimodal remote sensing image （RSI） fusion only perform simple fusion operations such as concatenation on the extracted multimodal features without designing dedicated modules for the integration of multimodal information， leading to insufficient fusion of complementary information across modalities. Secondly， these methods do not consider and utilize the cross-scale consistency priors within RSIs， resulting in limited extraction and integration of multimodal remote sensing information， and thus the performance of various downstream tasks needs to be improved. In response to the above issues， a multimodal RSI fusion method based on self-supervised pre-training and cross-scale contrastive learning is proposed， which mainly includes three parts： 1） by introducing a cross-attention fusion mechanism to preliminarily integrate features extracted from different modalities， and then using encoder modules to further extract features， explicit aggregation and extraction of complementary information from each modality are achieved； 2） by introducing a cross-modality fusion mechanism， each modality can extract useful supplementary information from the features of all modalities， and reconstruct each modality’s input after separate decoding； 3） based on the cross-scale consistency constraints of RSIs， cross-scale contrastive learning is introduced to enhance the extraction of single-modality information， achieving more robust pre-training. Experimental results on multiple public multimodal RSI fusion datasets demonstrate that， compared with existing methods， the proposed algorithm has achieved significant performance improvements in various downstream tasks. On the Globe230k dataset， our method achieves an average intersection over union （mIoU） of 79.01%， an overall accuracy （OA） of 92.56%， and an average F1 score （mF1） of 88.05%， and it has the advantages of good scalability and easy hyperparameter setting.

Abstract:In this paper， the small-signal modeling of the Indium Phosphide High Electron Mobility Transistor （InP HEMT） based on the Transformer neural network model is investigated. The AC S-parameters of the HEMT device are trained and validated using the Transformer model. In the proposed model， the eight-layer transformer encoders are connected in series and the encoder layer of each Transformer consists of the multi-head attention layer and the feed-forward neural network layer. The experimental results show that the measured and modeled S-parameters of the HEMT device match well in the frequency range of 0.5-40 GHz， with the errors versus frequency less than 1%. Compared with other models， good accuracy can be achieved to verify the effectiveness of the proposed model.

Abstract:A novel substrate integrated microstrip to ultra-thin cavity filter transition operating in the W-band is proposed in this letter. The structure is a new method of connecting microstrip circuits and waveguide filters， and this new structure enables a planar integrated transition from microstrip lines to ultra-thin cavity filters， thereby reducing the size of the transition structure and achieving miniaturization. The structure includes a conventional tapered microstrip transition structure， which guides the electromagnetic field from the microstrip line to the reduced-height dielectric-filled waveguide， and an air-filled matching cavity which is placed between the dielectric-filled waveguide and the ultra-thin cavity filter. The heights of the microstrip line， the dielectric-filled waveguide and the ultra-thin cavity filter are the same， enabling seamless integration within a planar radio-frequency （RF） circuit. To facilitate testing， mature finline transition structures are integrated at both ends of the microstrip line during fabrications. The simulation results of the fabricated microstrip to ultra-thin cavity filter transition with the finline transition structure， with a passband of 91.5-96.5 GHz， has an insertion loss of less than 1.9 dB and a return loss lower than -20 dB. And the whole structure has also been measured which achieves an insertion loss less than 2.6 dB and a return loss lower than -15 dB within the filter''s passband， including the additional insertion loss introduced by the finline transitions. Finally， a W-band compact up-conversion module is designed， and the test results show that after using the proposed structure， the module achieves 95 dBc suppression of the 84 GHz local oscillator. It is also demonstrated that the structure proposed in this letter achieves miniaturization of the system integration without compromising the filter performance.

Abstract:Phase consistency is produced under the requirements from the process of research， manufacture and realization in two-channel even multi-channel vector receivers. The required two-channel vector signals will be generated in this verification for stimulating the target systems or target sub systems， which can verify and emulate the performance in these systems or sub systems， and providing strong theories and technologies in the process of research and manufacture in systems. The specialized phase consistency measuring software controls general instruments and specialized instruments， which can be applied in the process of collocation， calibration， verification and signal generation for two-channel phase consistency signal system. The vector monitor and calibration for output network for specialized instrument have been studied and developed based on millimeter wave phase consistency， including many technologies for realization and research such as radio frequency amplitude-phase calibration， vector signal calibration， amplitude-phase measurement and restraint， and output signal threshold protection， meanwhile all testing parts are connected by Ethernet interface for feedback and control synchronization in the system.

Abstract:In this paper， a scheme of commonly-resonated extended interaction circuit system based on high order TM_n1 mode is proposed to lock the phases of two extended interaction oscillators （EIOs） for generating high power at G-band. Two separate EIOs are coupled through a specific single-gap coupling field supported by a designed gap waveguide with length L_g， which form the phase-locked EIOs based on the commonly-resonated system. As a whole system， the system has been focused on with mode analysis based on different single-gap coupling fields， mode hopping， which present the variation of phase difference between the two-beam-wave interactions when changing L_g. To demonstrate the effectiveness of the proposed circuit system in producing the phase locking， we conducted particle-in-cell （PIC） simulations to show that the interesting mode hopping occurs with the phase difference of 0 and π between the output signals from two output ports， corresponding to the excitation of the TM_n1 mode with different n. Simulation results show that 1） the oscillator can deliver two times of the output power obtained from one single oscillator at 220 GHz， 2） the two EIOs can still deliver output signals with phase difference of 0 and π when the currents of the two beams are different or the fabrication errors of the two EIO cavities are taken into account. The proposed scheme is promising in extending to phase locking between multiple EIOs， and generating higher power at millimeter-wave and higher frequencies.

Abstract:The photoconductive antenna is a very important device in the terahertz region and is widely used in terahertz time-domain spectroscopy technology. This article uses the molecular beam epitaxy method to grow Be-doped InGaAs/InAlAs superlattice as light absorbing materials for 1550 nm laser pumped photoconductive antenna for terahertz detection. The prepared materials have a sheet resistance greater than 10⁶ Ω/sq and an electron mobility of 216 cm²/（V?s）. The active mesa and electrode structure of the detection antenna are prepared using wet etching and magnetron sputtering processes， and the antenna chip is packaged on a PCB board. A detection antenna measurement system is built by employing a domestically produced 1550 nm femtosecond pump laser， and the detection antennas with electrode gaps of 40 μm and 60 μm are characterized. The measurement results indicate that the 60 μm antenna has a wider spectral width and power dynamic range， reaching 4.0 THz and 77.0 dB， respectively.

Abstract:Based on two-dimensional triangular lattice photonic crystal， a multi-functional device of two-dimensional photonic crystal waveguide light-emitting beaming with integrated filtering function is designed， which can achieve the fusion of outgoing light beaming and specific wavelength efficient filtering. The beam structure of the emitting light is similar to the grating structure， and the beam is in the state of clustering through the mutual interference principle between the multi-channel， which improves the radiation efficiency and distance of the emitting light. The finite difference time domain method can be used to obtain the effective propagation distance of 450 at the incident wavelength of 1.447 . The structure design has a good beaming ability for incident light in the range of 1.435-1.465 . At the same time， there are two hexagonal coupled filtering structures on either side of the waveguide， the transmission efficiencies for incident light waves are close to 98.4% and 97.3% with central wavelengths of 1.490 and 1.510 ， respectively. The fusion structure successfully realizes the filtering and clustering functions.

Abstract:A transceiver module operating at the 220 GHz frequency band was developed， consisting of three parts： a local oscillator chain， a transmitter chain， and a receiver chain， featuring high integration. A 218-226 GHz waveguide bandpass filter was designed to suppress spurious signals in the chain. The filter adopts a dual-mode resonant cavity structure to introduce a transmission zero on the left side of the passband， which suppresses the 214 GHz spurious signal by 60 dBc. An improved E-plane magic-T structure was used to form a four-way power combining amplifier to meet the requirement of transmit power. This module achieves a power combining efficiency of 72.5% and the output power is higher than 82 mW. The measured results show that in the 219.5-221 GHz frequency range， the transmit power is 82-95 mW， the noise figure of the receiver is less than 7.1 dB， the receiver gain is 5.1-6.0 dB， and the volume of module is 65×70×30 mm3.

Abstract:This study involved a comprehensive investigation aimed at achieving efficient multi-millijoule THz wave generation by exploiting the unique properties of cylindrical GaAs waveguides as effective mediators of the conversion of laser energy into THz waves. Through meticulous investigation， valuable insights into optimizing THz generation processes for practical applications were unearthed. By investigating Hertz potentials， an eigenvalue equation for the solutions of the guided modes （i.e.， eigenvalues） was found. The effects of various parameters， including the effective mode index and the laser pulse power， on the electric field components of THz radiation， including the fundamental TE （transverse electric） and TM （transverse magnetic） modes， were evaluated. By analyzing these factors， this research elucidated the nuanced mechanisms governing THz wave generation within cylindrical GaAs waveguides， paving the way for refined methodologies and enhanced efficiency. The significance of cylindrical GaAs waveguides extends beyond their roles as mere facilitators of THz generation； their design and fabrication hold the key to unlocking the potential for compact and portable THz systems. This transformative capability not only amplifies the efficiency of THz generation but also broadens the horizons of practical applications.

Abstract:When a gas leak occurs， it propagates through space in the form of diffusion， typically forming a gas plume with dynamically stable concentration near the leakage source， which appears as a quasi-static region in infrared images. This characteristic often causes reduced detection accuracy of conventional moving object detection algorithms in these regions and makes it difficult to obtain the spatial concentration distribution of the gas. To address this issue， a Vibe Gases adaptive threshold detection algorithm based on the background subtraction method is proposed， which introduces improvements in two critical phases of gas plume imaging. During the foreground extraction phase， a foreground difference matrix is first constructed through gas detection logic and subjected to two-dimensional frequency mapping. Subsequently， the optimal threshold for separating the foreground and background is calculated by fitting a difference distribution function using the least squares method. In the background updating phase， a signal matrix of the foreground gas is established and processed with two-dimensional frequency mapping. The primary signal range is then extracted through frequency-based high-pass filtering， followed by delayed updates for pixels located within both the gas region and this primary signal range. The experimental results of infrared detection imaging under stable gas leakage conditions demonstrate that at a distance of 20 meters， the detection accuracy for ethylene reaches 91.0% with an Intersection over Union （IoU） metric of 89.4%， while at 5 meters， the accuracy for detecting small leaks of sulfur hexafluoride is 81.3% with an IoU of 80.7%. The algorithm significantly improves the imaging quality of gas plumes， enhances adaptive detection capabilities across diverse gases and scenarios， and effectively extracts spatial concentration distributions of gases.

Abstract:This article proposes a three-dimensional light field reconstruction method based on neural radiation field （NeRF） called Infrared NeRF for low resolution thermal infrared scenes. Based on the characteristics of the low resolution thermal infrared imaging， various optimizations have been carried out to improve the speed and accuracy of thermal infrared 3D reconstruction. Firstly， inspired by Boltzmann''s law of thermal radiation， distance is incorporated into the NeRF model for the first time， resulting in a nonlinear propagation of a single ray and a more accurate description of the physical property that infrared radiation intensity decreases with increasing distance. Secondly， in terms of improving inference speed， based on the phenomenon of high and low frequency distribution of foreground and background in infrared images， a multi ray non-uniform light synthesis strategy is proposed to make the model pay more attention to foreground objects in the scene， reduce the distribution of light in the background， and significantly reduce training time without reducing accuracy. In addition， compared to visible light scenes， infrared images only have a single channel， so fewer network parameters are required. Experiments using the same training data and data filtering method showed that， compared to the original NeRF， the improved network achieved an average improvement of 13.8% and 4.62% in PSNR and SSIM， respectively， while an average decreases of 46% in LPIPS. And thanks to the optimization of network layers and data filtering methods， training only takes about 25% of the original method''s time to achieve convergence. Finally， for scenes with weak backgrounds， this article improves the inference speed of the model by 4-6 times compared to the original NeRF by limiting the query interval of the model.

Abstract:In order to meet the urgent need of infrared search and track applications for accurate identification and positioning of infrared guidance aircraft， an active-detection mid-wave infrared search and track system （ADMWIRSTS） based on "cat-eye effect" was developed. The ADMWIRSTS mainly consists of both a light beam control subsystem and an infrared search and track subsystem. The light beam control subsystem uses an integrated opto-mechanical two-dimensional pointing mirror to realize the control function of the azimuth and pitch directions of the system， which can cover the whole airspace range of 360°×90°. The infrared search and track subsystem uses two mid-wave infrared cooled 640×512 focal plane detectors for co-aperture beam expanding， infrared and illumination laser beam combining， infrared search， and two-stage track opto-mechanical design. In this work，the system integration design and structural finite-element analysis were conducted， the search imaging and two-stage track imaging for external scenes were performed， and the active-detection technologies were experimentally verified in the laboratory. The experimental investigation results show that the system can realize the infrared search and track imaging， and the accurate identification and positioning of the mid-wave infrared guidance， or infrared detection system through the echo of the illumination laser. The aforementioned work has important technical significance and practical application value for the development of compactly-integrated high-precision infrared search and track， and laser suppression system， and has broad application prospects in the protection of equipment， assets and infrastructures.