Proper total pressure, temperature and humidity environment in the cabin are the basic conditions for the stable operation of the space station platform and the safety of astronauts. For the long-term and reliable measurement of the total pressure, temperature and humidity environment in the cabin, an intelligent integrated monitoring technology and instrument are introduced, including the overall design, sensor design, integration challenges and calibration scheme. From the verification of on-orbit data, it is concluded that the instrument can achieve high-precision and high-reliability measurement of the total pressure, temperature and humidity in the cabin. It not only improves the level of environmental measurement in China's space station, but also provides new technical support for environmental detection of the manned space program in the future.

Aiming at the temperature measurement requirements of fiber grating sensors, this paper proposes and implements a Zynq-based Fiber Bragg Grating (FBG) temperature demodulation system. The system adopts the Zynq architecture, and the acquisition of FBG spectra and the calculation of the center wavelength are realized in the hardware system. Through the analysis and optimization of Gaussian fitting peak-seeking algorithm, the matrix operation is used to reduce the amount of curve-fitting operations, and the FBG center wavelength solving is realized in the hardware system, which takes into account the demodulation accuracy and at the same time, improves the demodulation rate and real-time performance of the system. The miniaturized fiber grating temperature demodulation system developed has a wavelength demodulation range of 1 510 nm ~ 1 590 nm with a stability of ±2 pm, which has a good potential application value in structural monitoring and other fields.

It is of great significance to obtain accurate heat flux data for understanding the thermal environment and optimizing the thermal protection system of the flight vehicles. The important aspects of heat flux measurement technology development are heat flux calibration and heat flux data processing. The heat flux sensor based on the hybrid heat transfer principle needs to utilize heat flux terms from different sources, such as the heat storage term and the temperature difference term to achieve heat flux results during data processing. The error sources and the magnitudes of influence of these terms are different, resulting in limitations of the common single-coefficient calibration method. Especially when the testing time is longer, some error factors will cause changes in heat flux measurement error. This paper proposes a multi-coefficient calibration method that corrects each heat flux term using different coefficients separately. Based on the temperature data in the calibration test and the known input heat flux, the calibration coefficient matrix is solved using the least squares method. The analysis and processing results of the calibration testing data show that using the multi-coefficient calibration method, the heat flux results are in better agreement with the input heat flux than those using the single-coefficient method. With enough testing data, the calibration deviation is less than 5% within the calibration time.

This study explores the combination of t-Distributed Stochastic Neighbor Embedding (t-SNE) dimensionality reduction technique and the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to address the challenges in multi-parameter radar signal sorting. As the complexity of radar signals has been increasing, traditional signal processing methods have revealed limitations. t-SNE effectively extracts essential features from the data by reducing dimensionality, eliminating noise and redundant information, and providing a clearer boundary for subsequent DBSCAN clustering. In the experiment, we generated five different types of radar signal data and conducted analyses using t-SNE and DBSCAN. The results show that the t-SNE dimensionality reduction combined with the DBSCAN clustering algorithm performs well in terms of purity and silhouette score, confirming the effectiveness of this method in complex radar signal sorting.

With the rapid development of 5G communication and millimeter-wave technology, the performance requirements for high-frequency passive interconnect structures in radio frequency (RF) microsystems have become increasingly critical. To address the efficiency bottlenecks caused by fragmented process data and isolated models in traditional design workflows, this paper proposed an independently developed 1~40 GHz silicon-based passive interconnect Process Design Kit (PDK). By integrating equivalent circuit models with HFSS full-wave electromagnetic simulation data, parameterized models for core structures such as grounded coplanar waveguide (GCPW) and micro-bump interconnects are established, and high-precision model matching is achieved through gradient optimization algorithms. The PDK development is completed on the Keysight ADS platform, including symbol libraries, parameterized cells, design rules, and verification workflows. Experimental results demonstrate that the developed PDK achieves a root mean square error (RMSE) of S-parameters below 10% across the 1~40 GHz frequency band. Based on this PDK, the simulation design of an X-band RF micro system was completed. The microsystem meets the specified performance requirements, verities the validity of the PDK. This PDK provides for reliable support for efficient design-process co-optimization in high-frequency RF systems.

In recent years, low-orbit navigation enhancement system has been gradually incorporated into the construction of integrated PNT system, and the capture and tracking of low-orbit signals has gradually become a hot research issue. In the scenario that the integrated PNT application terminal needs to receive BeiDou + LEO navigation augmentation signals at the same time, due to the large resource consumption of the capture module, using two sets of hardware resources to compatibly receive high, medium and low orbit-signals will cause a great waste of resources, and it is not possible to realize the small size and low power consumption requirements of the user terminal. This paper further optimizes the algorithm on the basis of the existing BeiDou signal capture module, and through the design of flexible and reusable matched filter plus FFT structure, on-chip distributed processing and off-chip ultra-long storage architecture, and techniques such as high Doppler pseudo-code, carrier frequency compensation, etc., it realizes the compatible reception of high, medium and low-orbit signals, and at the same time, it can realize the capture of high and medium orbit BeiDou B1C signals with a capture sensitivity of -145 dBm capture sensitivity and 40 kHz STL burst signal in low orbit.

With the increasing number of space debris and the increasing complexity of spacecraft missions as well as the requirements for adaptability to extreme environments, the operational status, damage diagnosis, life prediction and reliable safety as-sessment of spacecraft structure are particularly important. Operating as an effective approach, spacecraft structural health monitoring obtains structural feature information through sensor systems deployed in the structure，analyzes and evaluates the structural status through algorithm processing, thereby ensuring the safe and stable operation of the spacecraft at all stages. This paper focuses on the key technologies of spacecraft structural health monitoring. Firstly, from the sensor end of information acquisition, the technical characteristics, application status, current problems and development directions of optical fiber sensing, acoustic emission sensing and surface acoustic wave sensing are reviewed. Then, the research progress of sensor deployment methods and diagnostic evaluation algorithms for information processing is introduced. Secondly, the development trends and main challenges of spacecraft structural health monitoring are summarized and prospected.

To meet the application requirements of inertial navigation, autonomous driving, and other fields, and to advance the development of micro-electromechanical systems (MEMS) silicon gyroscopes towards high precision, digitalization, and miniaturization, this paper presents the design and implementation of a MEMS silicon gyroscope interface ASIC with digital output, based on a 0.35 μm BCD process and a monolithic integration approach. A closed-loop drive scheme based on noise self-excitation is adopted, enabling the gyroscope to achieve harmonic vibration in the drive direction. The detection circuit uses a low-noise capacitive-to-voltage (C/V) conversion circuit to efficiently convert the small displacement signal into a voltage signal. Signal processing is performed using switched-capacitor phase-sensitive demodulation technology, combined with low-pass filtering, effectively suppressing noise interference and yielding a low-noise analog angular velocity output signal. To achieve the digital output of the silicon gyroscope's angular velocity, an integrated fourth-order feed-forward Sigma-Delta (ΣΔ) analog-to-digital converter (ADC) is designed to convert the analog angular velocity signal into a digital signal. The chip test results show that the dynamic range of the ΣΔ modulator reaches 110 dB, with a low-frequency noise floor of approximately -120 dB. The overall range of the gyroscope is ±200(°)/s, with a scale factor of 21 310 LSB/((°)/s), a nonlinearity of 178×10^-6, bias instability of 0.259(°)/h, and angle random walk of 0.028 7(°)/√h. The chip area is 4.3 mm×4.3 mm. By using the integrated interface ASIC to replace the traditional PCB-level system, the system's integration is significantly improved, successfully meeting the miniaturization requirements for MEMS silicon gyroscopes and promot-ing their development in high-precision digital applications.

Micro capacitive pressure sensors have important application value in fields such as biomedicine, drone positioning, and wearable devices. This study focuses on MEMS capacitive gauge pressure sensors. A high-precision capacitive pressure sensor is designed and fabricated for implantable biomedical applications. This device is formed by anodic bonding between a glass substrate with fixed electrodes and an SOI wafer with elastic membranes. The elastic membrane will deform when the pressure on both sides changes, resulting in a change in the capacitance of the device. The readout circuit calculates the external pressure by detecting changes in capacitance. This paper improved the nonlinearity of capacitive pressure sensors by designing a structure of a boss on the membrane. The result from ANSYS finite element simulation shows that the nonlinearity was improved from about 17% (flat membrane) to 7% (with boss structure). The structure of the boss was formed by anisotropic etching of silicon with TMAH, and the SOI buried oxide layer serves as the stop layer of etching to achieve high device consistency. Finally, a packaging and testing platform is built for capacitive pressure sensors and calibration. In the measurement range of 0 ~ 40kPa relative to atmospheric pressure, this device achieved a testing accuracy of 0.30%, nonlinearity of 8%, and repeatability error as low as 0.09%.

In response to the demand for in-situ continuous measurement of molten steel temperature in the field of metallurgy, a novel contact-type continuous temperature measurement sensor structure scheme is proposed. The thermal response characteristics are analyzed by establishing a three-dimensional heat transfer model. Then a high-pressure process and a high-temperature annealing process are used to realize the preparation of an integrated sensitive component of thermocouple/ceramic protection body, which is then encapsulated inside the lance-type support structure of MgO-C material for making a temperature measurement sensor; Continuous temperature measurement tests are carried out under the environment of molten steel of Ladle Furnace, and the results show that in the 1 600 ℃~1 650 ℃ high temperature molten steel environment, the sensor can effectively measure temperature continuously for more than 16 minutes.