Semiconductor Optoelectronics
Co-Editors-in-Chief
Changlin Liu
2025
Volume: 46 Issue 2
27 Article(s)

Sep. 18, 2025
  • Vol. 46 Issue 2 1 (2025)
  • TIAN Huili, YANG Zhangyi, GUI Wenzhu, and DONG Wei

    This paper proposes a high-sensitivity fiber magnetic field sensor based on a Mach-Zehnder interferometer and a photoelectric oscillator. According to the principle of the sensor based on the photoelectric oscillator, the relationship between the wavelength sensitivity of the designed Mach-Zehnder interferometer and the frequency sensitivity of the photoelectric oscillator is established, revealing that the frequency sensitivity of the photoelectric oscillator increases as the wavelength sensitivity increases. In this design, the Mach-Zehnder interferometer is constructed by splicing a tapered fiber between two single-mode fibers with an offset, which can split the light emitted by a broadband light source into sinusoidal-shaped waves. Experimental results show that within the magnetic field range of 1.6 mT to 12.8 mT, the frequency sensitivity of the sensor is 324 Hz/mT, with a corresponding measurement error within ±0.96 mT.

    Sep. 18, 2025
  • Vol. 46 Issue 2 189 (2025)
  • SHI Yu, WEI Zecheng, ZHANG Rong, and KONG Mei

    Optical-fiber-localized surface plasmon resonance (LSPR) sensing offers the advantages of label-free detection, low sample consumption, and long-distance sensing. In this study, using the liquid-liquid interface self-assembly method, a uniform and dense gold trisoctahedral monolayer film is rapidly assembled on the surface of an optical fiber, following which an optical-fiber LSPR sensor is successfully prepared. Characterization test results show that gold trisoctahedra are regularly arranged and uniformly distributed without agglomeration, and that the surface of the optical fiber is well coated. Simulation analysis shows that compared with gold nanospheres of the same size, the multitipped structure of a gold trisoctahedron offers a more significant local electric field enhancement effect. Combined with microfluidic technology, the refractive index of sucrose solutions at varying concentrations is successfully sensed by the optical-fiber LSPR sensor, achieving a maximum sensitivity of 312 nm/RIU and demonstrating favorable detection performance.

    Sep. 18, 2025
  • Vol. 46 Issue 2 195 (2025)
  • ZHANG Xuan, ZHANG Chunming, and SONG Yidi

    With the continuous development of information technology, communication systems require high data transmission rates. To extend the high-frequency bandwidth and reduce the circuit power consumption simultaneously, we designed a 60 Gb/s PAM4 signal SerDes receiver for receiving quarter-rate architecture. A continuous time linear equalizer (CTLE) with transconductance trans-impedance amplifier (Gm-TIA) structure and a parallel four-tap feed forward equalizer (4-tap FFE) equalization strategy are proposed to extend the high-frequency bandwidth, and a quarter-rate structure is used to further reduce the circuit power consumption. The circuit is implemented based on a 28 nm CMOS process at 1.05 V/1.2 V supply voltage, and post-simulation results show that the CTLE compensates for 18.5 dB of channel fading at Nyquist's 15 GHz frequency, and after a 4-tap FFE eye width of 0.6 UI and an eye height of 120 mV, the overall receiver-side power consumption is 83.71 mW, with an energy efficiency of 1.40 pJ/bit. The proposed circuit has a core area of 394.55 μm × 343.53 μm and is suitable for high-speed serial interface and communication applications.

    Sep. 18, 2025
  • Vol. 46 Issue 2 202 (2025)
  • WANG Chongxiao, CHEN Weijun, SONG De, LIANG Rongxuan, YUE Jipeng, XIA Haoran, and LIU Mingshan

    Based on photon-transport theory and the theory of electron-semiconductor interactions, a model for the formation mechanism of gain noise caused by secondary electron fluctuations in the electron-multiplying layer of electron-bombarded complementary metal-oxide-semiconductor (EBCMOS) devices is established. By performing a Monte Carlo simulation, the signal-to-noise ratio (SNR) of secondary electrons is calculated to analyze the effect of system parameters, including incident electron energy, passivation-layer type and thickness, electron-multiplying layer thickness, and doping concentration, on gain noise. The results indicate that by using Al2O3 as the passivation-layer material and implementing a series of measures such as reducing the passivation-layer thickness, doping concentration, electron-multiplying layer thickness, and operating temperature, the SNR of the electron-multiplying layer can be effectively improved to 75.35 dB. The findings of this study provide theoretical guidance for preparing the electron-multiplying layer in EBCMOS devices.

    Sep. 18, 2025
  • Vol. 46 Issue 2 209 (2025)
  • WANG Peiyan, CHEN Weiqiang, LU Lidan, ZHU Lianqing, and ZHANG Dongliang

    To satisfy the demand for linear avalanche photodetectors (APDs) in low-light detection, an APD based on an InAs/GaSb second-class superlattice and AlGaAsSb quaternary alloy as absorption and multiplier layers, respectively, was designed in this study. By constructing a collision ionization model for the AlGaAsSb alloy and using the SILVACO-ATLAS simulation platform, the effects of structural parameters such as the thickness and doping concentration of the multiplication and absorption layers on the device's I–V characteristics, electric-field distribution, and transient response were systematically simulated. The aim was to obtain the optimal model parameters to reduce dark current and improve gain. Additionally, the effect of different temperatures on device performance was investigated. Simulation results show that at 300 K, the gain at the operating voltage is 16.4 and the maximum gain at the breakdown voltage is 129.5, with a specific detectivity of 2.39 × 1011 cm·Hz1/2·W−1. This study provides a theoretical basis for the development of medium-wave infrared avalanche photodiodes operating at room temperature.

    Sep. 18, 2025
  • Vol. 46 Issue 2 216 (2025)
  • ZHANG Tengfei, WANG Wei, and HUA Wenyuan

    In this study, the current-voltage characteristics and current-noise power spectral density of GaN-based blue micro-LED chips are examined to investigate the current-transport mechanism and low-frequency noise behavior under different current injections. The results show that under low current injection (1 nA<I<1 A), the low-frequency noise is primarily a superposition of G-R noise and noise, which occurs at a frequency of less than 10 Hz. The current noise slopes are 1.75, 2.06, 1.99, and 1.76 at intermediate current injections (6.6 A<I<830 A) and at a frequency of below 1×103 Hz. The noise-signal behavior is investigated based on the multiscale entropy complexity, which reflects the noise-signal irregularity, and suggests that the current noise under this current-injection range originates from electron and hole generation-complexation instead of from defect-assisted tunneling. Under high current injection (1 mA≤I≤10 mA), as the current increases, the series resistance is not negligible and the current no longer increases exponentially with the increase in voltage. Thus, the noise 1/f primarily originates from the series resistance and corresponds to noise exponent factors of 1.53, 1.42, 1.37, 1.24, and 1.03, which gradually approach that of standard 1/f noise (=1).

    Sep. 18, 2025
  • Vol. 46 Issue 2 226 (2025)
  • GE Xiaolong, LI Zehao, LI Chunyan, JIANG Qi, ZHOU Hongyu, and CHANG Yuchun

    An EVS readout circuit was developed using the XFAB 180 nm CMOS process. The circuit was designed for an event-based vision sensor, analyzing the asynchronous timing of row and column scanners and the impact of event output data volume on circuit performance. Asynchronous timing is implemented through row and column scanning, improving time resolution and reducing the power consumption and delay of the readout circuit. The circuit automatically switches data output methods based on row event density, reducing the output of redundant event information and increasing the event readout rate. Each pixel integrates a logarithmic light sensing circuit, a switch-capacitor amplifier circuit, and a two-transistor comparator to generate event information, with a power consumption of 54.3 nW per pixel. The readout circuit operates at 1.8 V, with a maximum event rate of 41 Meps. In a 64×64 array, the total power consumption of the circuit at an event rate of 1 Meps is 15.31 mW, and the data output switching process is capable of reducing redundant event output by up to 50%.

    Sep. 18, 2025
  • Vol. 46 Issue 2 232 (2025)
  • WANG Jiawen, CUI Yiqian, YANG Xiaoqian, ZHANG Yu, and YU Wei

    This study systematically examines poly(3,4-ethylenedioxythiophene)∶polystyrene sulfonate (PEDOT∶PSS) and molybdenum oxide (MoOx) as hole transport materials for n-Si heterojunction tunnel oxide passivated contact (TOPCon) solar cells. Initially, the photoelectric properties of PEDOT∶PSS and MoOx films, including work function, resistivity, light absorption, and anti-reflection effect on silicon wafers, were analyzed. Subsequently, full-back electrode PEDOT∶PSS/n-Si and MoOx/n-Si heterojunction TOPCon solar cells, featuring rear-located P-N junctions, were fabricated and subjected to experimental comparisons. Performance evaluations, conducted using quantum efficiency (EQE), light J-V, dark J-V, and stability tests, indicate that PEDOT∶PSS/n-Si solar cells exhibited superior P-N junction quality and conversion efficiency. Conversely, MoOx/n-Si solar cells demonstrated higher open-circuit voltage and enhanced stability. These findings provide valuable insights for developing targeted optimization strategies to improve the performance of non-doped silicon-based heterojunction solar cells.

    Sep. 18, 2025
  • Vol. 46 Issue 2 240 (2025)
  • CHENG Ganlin, ZHANG Ruimeng, WANG Dan, MU Shuaichen, and FAN Wei

    The reliability of image sensors, as core components of remote-sensing payloads, directly determines the overall lifespan of remote-sensing satellites. CMOS image sensors have become the primary visible-light detectors for remote-sensing payloads. However, the existing lifetime - evaluation methods rely primarily on quality-assurance tests, which are devoid of quantitative analysis. This study proposes a mission requirement-based lifetime-evaluation method for space-applied CMOS image sensors. Based on a virtual remote-sensing satellite mission as a case study, a reliability analysis of the device (including failure mechanisms such as the time-dependent dielectric breakdown (TDDB)) was conducted at the design stage, and a quality-assurance scheme was formulated during the test phase. Through irradiation experiments focusing on the performance degradation caused by the total ionizing dose, displacement damage, and TDDB effects as well as predictions of cumulative irradiation levels in the mission environment, key indicators such as the imaging signal-to-noise ratio (SNR) of the CMOS image sensor at the end of the camera's lifespan were quantitatively evaluated. Experimental results indicate that the performance degradation of the device remains controllable within the 8-year design lifetime, with the SNR satisfying the mission requirements. This method provides theoretical support for the reliability design and verification of long-life remote-sensing cameras.

    Sep. 18, 2025
  • Vol. 46 Issue 2 248 (2025)
  • LI Yao, GOU Henglu, CHEN Lihuan, NIU Ruixia, CHEN Wenshu, LONG Yi, and BAI Kailiang

    In this study, a deep trench (DT) semi-superjunction (SSJ) vertical double-diffused metal-oxide semiconductor (VDMOS) field-effect transistor was developed. Silvaco TCAD simulation software was used to simulate and investigate the electrical characteristics of the DT SSJ VDMOS, which were compared with that of the DT VDMOS and DT super-junction (SJ) VDMOS. The DT overcomes the limitation of the Gaussian law of the vertical electric field at the gate oxide interface, and reduces the peak electric field of the gate oxide layer. The introduction of the SSJ reduces process complexity and simultaneously improves device stability compared with those of the SJ structure. The simulation results show that the breakdown voltage (BV), specific on-resistance (Ron, sp), and figure of merit (FOM) of the DT SSJ VDMOS are 1 000 V, 26.78 m· cm2, and 37.565 MW/cm2, respectively. The BV of the DT SSJ VDMOS device increased by 100%, and FOM by 57% compared to those of the DT VDMOS device. Furthermore, its Ron, sp was reduced by 35.9%, and FOM was increased by 5% compared with those of the DT SJ VDMOS device. By adopting DT and SSJ structures, the contradiction between the breakdown voltage and Ron, sp is resolved, and the overall performance of the device is improved.

    Sep. 18, 2025
  • Vol. 46 Issue 2 255 (2025)
  • LI Zhenqiu, ZHONG Wansheng, SHU Lincong, LI Shan, ZHANG Jiahan, and LIU Zeng

    Doping can significantly affect the optical absorption characteristics and electrical properties of Ga2O3 via bandgap modulation, thus improving the performance of photodetectors. This study uses plasma-enhanced chemical vapor deposition (PECVD) to fabricate In-doped Ga2O3 films and investigates the influence mechanism and application potential of In doping on the performance of -Ga2O3-based solar-blind photodetectors. This photodetector demonstrates remarkable deep-ultraviolet (DUV) response properties, reaching a photocurrent level of microamperes under 5 V bias and 275 W/cm2 illumination at a wavelength of 254 nm. Additionally, it achieves a maximum responsivity of 480 mA/W, a detectivity of 6.42×1013 Jones, an external quantum efficiency of 235%, a photoconductive gain of 2.15, and a maximum photo-to-dark current ratio of approximately 1×106, thus demonstrating the extremely high sensitivity of In-doped Ga2O3 to low DUV light intensities and establishing its significant potential in optical sensing applications. High-temperature growth facilitates In doping; therefore, the photocurrent of In-doped -Ga2O3 films is significantly higher than that of their counterparts without In doping under identical illumination and bias conditions. The presence of oxygen vacancies and lattice defects in the films significantly affects the performance of the photodetector, with surface oxygen vacancies affecting electron accumulation and migration through the adsorption of oxygen molecules and the formation of surface barriers. Consequently, the photocurrent response is enhanced under 254 nm ultraviolet light, whereas the optical-response parameters diminish as the light intensity increases, which is primarily due to the scattering and recombination of charge carriers. This study provides a comprehensive discussion on the fabrication methods and photoelectronic characteristics of In-doped Ga2O3 films. Our findings offer new insights into the design of high-efficiency In-doped Ga2O3 DUV photodetectors and provide new avenues for the application of PECVD technology in semiconductor materials.

    Sep. 18, 2025
  • Vol. 46 Issue 2 261 (2025)
  • ZHONG Guoshun, LI Hongyu, HU Guohua, and PENG Song

    A coarse-wavelength-division multiplexer (CWDM) chip, based on grating-assisted contra-directional couplers, was designed, fabricated, and experimentally characterized. The device integrates three gratings, with distinct reflection wavelengths, in a cascaded configuration. The grating-assisted contra-directional coupler structure eliminates the need for discrete circulators as well as effectively mitigates the additional 6-dB loss typically associated with traditional on-chip beam splitters. Furthermore, this design remarkably suppresses crosstalk between the reflected light and output signal. Leveraging a silicon-on-insulator platform, key parameters such as grating width and the number of periods were optimized, and an apodization design methodology was adopted for performance enhancement. Consequently, a compact three-channel CWDM chip was successfully fabricated. Experimental measurements indicate that the on-chip insertion loss and channel crosstalk are approximately 2.5 dB and below &#x2212;24 dB, respectively, and the spectral response is characterized by a box-like shape. The chip occupies an area of merely 0.6 mm × 1.5 mm. This study addresses the critical technical challenges related to large device dimensions, narrow bandwidths, and the realization of box-like spectral responses in conventional CWDM devices. The developed chip is promising for metropolitan access networks, fiber-to-the-home systems, data center interconnections, and silicon-based photonic integrated circuits.

    Sep. 18, 2025
  • Vol. 46 Issue 2 269 (2025)
  • ZHANG Zhizi, WANG Zicheng, KUANG Chongwei, ZHU Changsheng, and ZHANG Xiangjun

    An integrated interferometric fiber optic gyroscope (I-IFOG) is a promising solution for next-generation interferometric fiber optic gyroscopes (IFOG) owing to its miniaturization and low power consumption. After integrating the optical components onto a chip, a significant polarization error arises because of the absence of polarization-maintaining fiber connections, which significantly reduces the angular velocity measurement accuracy and gyroscope stability. The polarization error greatly restricts further use of the I-IFOG. To analyze the optical path polarization, a polarization error model of the integrated optical chip was established based on a Jones matrix. Using the mathematical model, we have found that the primary factors affecting polarization are the nonreciprocity noise induced by polarization crosstalk and mode dispersion. Therefore, a novel waveguide polarizer is proposed to remove the undesired polarized light and maintain a high degree of polarization. To verify the polarizer performance, the polarization state of the signal was measured using a polarization analyzer. The degree of polarization improved by 7.16% and the extinction ratio improved by 39.8% compared with those of a traditional polarizer. Finally, a prototype based on the referred polarization suppression method was generated to measure the Earth's speed, and the static performance was checked and processed. The accuracy of the prototype reached 0.112 9 (&#x00B0;)/h.

    Sep. 18, 2025
  • Vol. 46 Issue 2 275 (2025)
  • SHAO Boyou, WANG Kai, LEI Cheng, GONG Mingfeng, JI Diyang, ZHANG Zhenyu, GUO Jinzhu, and LIANG Ting

    Microelectromechanical system thermopile infrared detectors are an important infrared detection technology that is widely used in diverse fields. Owing to the continuous development of infrared technology and the increasing demand for its applications, the performance index for thermopile infrared detectors has increased. A thermopile structure with an integrated super-surface structure in the absorption region is designed to increase the absorption rate of infrared light from 9 to 11 m to about 70%. Additionally, the thermocouple strip is designed as a quadrilateral structure with a narrow hot end and a wide cold end, which allows the hot end to extend to the temperature-concentration region, thus reducing the internal resistance. Finite-element simulation software is used to simulate the structure, and the results show that the response voltage of the detector after integrating the super-surface can reach 37.876 mV. Meanwhile, the response and detection rates are calculated to be 41.9 V/W and 1.97×108 cm·Hz1/2·W&#x2212;1, respectively, with performance enhancement by 38.9%, 38.7%, and 39.7% recorded.

    Sep. 18, 2025
  • Vol. 46 Issue 2 284 (2025)
  • MAO Dongchen, ZHANG Wenjie, ZHAO Kangcheng, and ZHU Hao

    Monocrystalline silicon, a basic material for semiconductor devices, is characterized by high hardness, high brittleness, and low fracture toughness, which pose significant challenges to micro-nano processing. Laser machining is widely used to fabricate monocrystalline silicon microstructures, despite the accompanying notable thermal damage. To address this issue, a novel method is proposed for the through-groove machining of silicon wafers based on laser-electrochemical hybrid processing. This method leverages the photothermal coupling effect of lasers and increased conductivity of monocrystalline silicon with temperature, to achieve a combination of laser etching and electrochemical dissolution, thereby effectively improving the machining quality. The mechanism of laser-induced electrochemical machining is elucidated, and a corresponding experimental scheme is designed. Comparative experimental results demonstrated that laser-electrochemical hybrid processing surpasses traditional laser dry cutting methods in terms of surface quality. Further investigations into the effect of the applied voltage on machining surface quality revealed that at a voltage of 40 V, the surface roughness was significantly reduced to 310 nm, achieving optimal machining outcomes.

    Sep. 18, 2025
  • Vol. 46 Issue 2 293 (2025)
  • WANG Jianjian, TANG Rui, and SHANG Qiufeng

    To improve the transmission capacity of optical fiber communication, few-mode fibers that can realize mode division multiplexing have become a popular research topic, where the analysis of their bending characteristics is particularly significant. In this paper, the effects of bending radius and core radius on the bending characteristics of ordinary step few-mode fibers, ring-core few-mode fibers, and elliptical ring-core few-mode fibers are analyzed using finite element modeling. Through the comparative analysis of the three performance indexes (effective refractive index, bending loss, and effective mode field area), bending loss was determined to have less influence on the effective refractive index of the three types of few-mode fibers in each order mode. With the increase in bending radius, the bending loss of the few-mode fiber tended to decrease, while the effective mode field area tended to increase. The ring-core few-mode fiber had a higher refractive index difference owing to the special core structure, and the overlap between modes of elliptical ring-core few-mode fiber was minimal. This research provides a useful reference for the structural design and optimization of few-mode fibers as well as a theoretical basis for further research on the use of few-mode fibers to achieve high-quality signal transmission.

    Sep. 18, 2025
  • Vol. 46 Issue 2 299 (2025)
  • WANG Shuqi, and CHEN Jiale

    An ultrawide-band circularly polarized microstrip antenna loaded with arc-shaped branches was proposed for satellite communication in integrated sensing and communication. By cutting the square loop antenna and adjusting the surface transverse and longitudinal currents, circularly polarized waves were generated. The axial ratio bandwidth was expanded by loading two quarter arc-shaped branches. By designing an adjacent coupled feed structure that resonated with the top layer of the antenna, antenna miniaturization and impedance bandwidth expansion were achieved. The simulation and experimental results show that the antenna emits right-handed circularly polarized radiation; the impedance bandwidth is 9.06~22.55 GHz (85.4%); the 3 dB circular polarization axis ratio bandwidth is 9.91~21.54 GHz (74%); the peak gain is 3.5 dBi; and the dimensions of the antenna are 11.7 mm×11.7 mm×0.127 mm (0.38 mm×0.38 mm×0.127 mm, where is the free space wavelength corresponding to 9.91 GHz). The designed antenna structure is simple, compact, and has a wide operating frequency that can cover the Ku-band.

    Sep. 18, 2025
  • Vol. 46 Issue 2 307 (2025)
  • LUO Guoping, WANG Guohao, LI Wei, WANG Meizhen, JIANG Jingwen, LUO Yuanxing, and ZHU Weiling

    Oxide/metal/oxide multilayer-structured transparent electrodes are considered to be a promising alternative to indium tin oxide. In this research, ultra-thin metal-based transparent electrodes were deposited by a magnetron sputtering system to investigate the effect of high-vacuum annealing treatment on the microstructure, surface morphology, and optoelectronic properties of multilayer transparent electrodes. The experimental results showed that the high-vacuum annealing treatment affects the grain size and surface morphology of the films, which become denser, with lower losses to carrier scattering and improved optoelectronic performance of the transparent electrodes. After annealing at 200°C for 1 h under high vacuum, the ultra-thin metal-based transparent electrodes has better optical and electrical properties. The sheet resistance is 8.34 /□, the average visible transmittance is nearly 96%, and the figure of merit is as high as 1 090.45/. The results show that ultra-thin metal-based transparent electrodes have good thermal stability and excellent optoelectronic performance.

    Sep. 18, 2025
  • Vol. 46 Issue 2 313 (2025)
  • FENG Wei, YU Wenbo, LUAN Chao, XIANG Yuyan, and LI Song

    To meet the demand for long-distance and high-precision delay in the echo simulation equipment of satellite-mounted laser altimetry systems, a digital pulse delay generator (DDG) architecture, based on the internal integration of special delay lines in field-programmable gate arrays (FPGAs), is proposed. A self-measurement compensation-type DDG consists of “interpolation delay line + three levels of coarse and fine time-delay circuits.” The CARRY4 interpolated delay line is leveraged to accurately measure the time difference between the arrival of the external trigger signal and the FPGA internal-clock-acquisition edge and thus locate the starting point of the time delay with high accuracy. Then, the three-level time delay scheme of coarse time counting is applied for sequential coarse and fine phase adjustments—this approach enables the realization of high-precision delay on long-time scales based on the starting moment of the trigger signal. The DDG design incorporates jitter self-measurement and calibration methods, which are applicable to both internal and external triggering. In addition, its design resources rely on FPGA on-chip-based resources, ensuring wide adaptability of the design. Experimental results show that when an oscillator with a frequency accuracy of ±0.5 ppm is used, a minimum-time-delay step of up to 37.5 ± 7.5 ps is obtained in the compensated externally triggered mode, and for a 2-ms delay, the output jitter is better than 270 ps (the root mean square error is 31 ps).

    Sep. 18, 2025
  • Vol. 46 Issue 2 318 (2025)
  • WANG Jiaqiang, QIU Haitao, HE Yang, and SHI Haiyang

    The development of high-precision fiber optic gyroscope (FOG) shows a distinct trend toward system integration and structural miniaturization. The primary objective is to achieve device miniaturization through structural optimization while maintaining high measurement accuracy. In this study, the ADC, FPGA, and DAC dies of a FOG signal processing circuit were integrated onto a single substrate using gold wire bonding technology within a system-in-package (SiP) framework. To address the substantial impact of the physical parameters of gold bonding wires (e.g., wire diameter, span, and arc height) on signal integrity, a high-fidelity electromagnetic simulation model was developed to systematically investigate the coupling effects between the geometric parameters of gold bonding wires and the width of substrate microstrip lines on signal transmission performance. An L9(34) orthogonal experimental design was employed to conduct multi-objective optimization across four groups of critical parameter combinations, yielding an optimal configuration of SiP bonding wires and substrate microstrip lines tailored for FOG signal processing circuits. These findings provide practical guidance for SiP integration in FOG signal processing circuits and valuable insights for engineering applications.

    Sep. 18, 2025
  • Vol. 46 Issue 2 327 (2025)
  • YUAN Jianguo, YANG Ting, FU Bowen, and XIONG Longyu

    Short-cycle structures in quasi-cyclic low-density parity-check (QC-LDPC) codes can significantly degrade error correction. Thus, a novel construction method using QC-LDPC codes with girth-8 based on the search algorithm of the subtraction construction (SC) is proposed in this study. This method employs the search algorithm of the SC to design QC-LDPC codes with a girth of at least 8. In the construction method, the first row is first fixed based on the Golomb Ruler sequence, whereas the subsequent two rows are derived using the search algorithm of the SC. This results in the formation of an exponent matrix that satisfies the condition without girth-4 and girth-6, thus yielding the parity-check matrix. Simulation results show that, compared with other QC-LDPC codes of the same code rate and length, the constructed SC-QC-LDPC code can improve the minimum net coding gains of 0.23 and 0.12 dB for code lengths of 1 200 and 3 600, respectively, at a bit error rate of 10&#x2212;6. This shows that the proposed construction method performs better at improving the error-correction performance. Furthermore, it allows the selection of various code lengths and rates while maintaining low computational complexity.

    Sep. 18, 2025
  • Vol. 46 Issue 2 336 (2025)
  • SONG Jian, HUANG Jixun, FENG Wenshuai, and LIU Yuanyuan

    Erbium-doped superfluorescent fiber source is widely used in medium- and high-precision fiber-optic gyroscopes. The mean wavelength of the output light drifts as its use time increases. Consequently, the scale-factor stability of fiber-optic gyroscopes is insufficient to satisfy the requirements for long-term application. To solve this problem, this paper proposes a technology to improve the long-term stability of the mean wavelength, which involves incorporating a pump-power control loop to the light source to stabilize the pump power and reduce the pump-power loss and mean-wavelength drift caused by the aging of the pump source. Optical-path filtering technology is adopted to filter the unstable band light in the spectrum to improve the spectral stability and symmetry, and the filter parameters are optimized via simulation. Experimental results show that the mean-wavelength stability of the fiber-optic gyroscope using this method improves by 88%, the scale-factor stability exceeds 5 ppm within 9 months at room temperature, and other parameters of the improved fiber-optic gyroscope are not significantly degraded, thus demonstrating the effectiveness of the method.

    Sep. 18, 2025
  • Vol. 46 Issue 2 342 (2025)
  • HUANG Junyi, HE Wei, and LI Shaode

    In this study, a bolt stress test method based on Fiber Bragg Grating (FBG) sensing is proposed to solve the problems of loosening and creep of bolts in long-term service. First, a femtosecond-laser processing technology was used to fabricate an FBG through fiber coating, and an erbium-doped fiber laser with a linear cavity structure was designed based on the fiber grating to achieve a laser output with a high signal-to-noise ratio. Second, bolt gaskets of different sizes made of stainless steel and aluminum were designed and fabricated, and their mechanical characteristics were simulated and analyzed. Next, the FBG was embedded in the stainless steel and aluminum gasket grooves and combined with the bolts to build a stress test system. Applying stress to the surface of the gasket deformed the FBG was accordingly, causing the laser wavelength to drift. In the experiment, stress of 0~10 N·m was applied to the bolt strain test system made of stainless steel and aluminum gasket with groove depths of 1 and 1.5 mm. As the stress increased, the laser wavelength showed a red shift, and the center wavelength of the stainless steel and aluminum washer sensors with groove depths of 1 mm moved by 41 and 61.8 pm, respectively; their sensitivities were 4.05 and 6.13 pm/(N·m), respectively. The 1.5 mm groove-depth stainless steel and aluminum gasket sensors showed wavelength drifts of 141.7 and 180.5 pm, respectively, and sensitivities of 14.12 and 18 pm/(N·m), respectively, along with a linearity of 0.99.

    Sep. 18, 2025
  • Vol. 46 Issue 2 350 (2025)
  • ZHANG Weigang, XIE Zhihua, CHE Chi, HU Xiaodong, FU Zhengquan, and ZENG Xingyu

    To address the scarcity of large-pixel large-area array detector chips used in star sensors, a design scheme for an imaging module based on domestic high-sensitivity CMOS image sensors (CISs) is proposed. The hardware circuit of the CIS-based star-sensitive imaging module is designed based on the latter's operating principle. An interactive high-speed data-transmission mode for FPGA and DSP software is developed based on an advanced extensible interface bus-control module. Technical parameters related to the practical application of star sensors, i.e., non-uniformity, linearity, dynamic range, and signal-to-noise ratio, are tested. The results show that these indicators of the CIS star-sensitive imaging module satisfy the application requirements for star detection.

    Sep. 18, 2025
  • Vol. 46 Issue 2 360 (2025)
  • BAI Hongyu, YANG Shuai, DU Jiang, CHEN Long, ZHAO Benyuan, and MING Jia

    Misoperation of power grids is one of the main factors leading to equipment damage and system failure incidents. To enhance the grid′s ability to prevent misoperation, this paper proposes a multimodal data fusion-based power grid misoperation diagnosis method. First, video surveillance, sensor data, and operation logs are fused using a bidirectional gated recurrent unit (BiGRU). A multi-interaction attention mechanism is employed to obtain a unified representation of the multimodal data. Then, a Bayesian neural network is used to update the weights of the long short-term memory (LSTM) network. By combining the uncertainty estimation capability of Bayesian methods with the strength of LSTM in handling time-series data, the fused multimodal features are fed into a Bayesian LSTM network to diagnose grid misoperation. Finally, the simulation results show that the proposed method effectively improves the accuracy of misoperation diagnosis and further enhances the safety and stability of grid operations.

    Sep. 18, 2025
  • Vol. 46 Issue 2 367 (2025)
  • YANG Duo, LIAO Jingjing, LI Qihui, YANG Zhongxu, and XIN Jingtao

    To extend the operational range of the fiber grating sensing system into the near-infrared band accessible to silicon-based detectors, this study applies a phase mask to the 1 550 nm waveband for second-order Bragg grating inscription research on the 780 nm waveband. First, the formation mechanism of the second-order grating is analyzed, and then a grating-inscribing system is integrated with an ultraviolet laser equipped with a phase mask grating to fabricate the second-order FBG. The ratio of the first-order wavelength to the second-order wavelength was found to be 1.98, and the bandwidth of the second-order wavelength is narrower. The results indicate that the temperature and strain sensitivities of the second-order grating are 5.28 pm/℃ and 0.65 pm/, respectively, or approximately half the first order grating. This research provides technical support for expanding the range of the optical fiber sensing system and the waveband of the silicon-based detector.

    Sep. 18, 2025
  • Vol. 46 Issue 2 374 (2025)
  • Please enter the answer below before you can view the full text.
    Submit