This PDF file contains the front matter associated with SPIE Proceedings Volume 13495, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

With the continuous development of modern industry, high-precision dynamic monitoring systems are essential to ensure the safety and efficiency of production. Therefore, the terahertz radar system based on frequency modulation continuous wave (FMCW) was studied in this work, aiming at realizing high-precision dynamic monitoring of coal levels in coal bunkers. The system adopts the FMCW method to measure the target distance by transmitting continuous linear frequency modulation signals and using the frequency difference and time delay between the received signal and the transmitted signal. The frequency selection algorithm is combined with the phase estimation algorithm, and the spectrum is refined by Zoom-FFT algorithm to improve the range resolution. The phase estimation algorithm is used to supplement and improve the ranging accuracy to ensure the reliability of the ranging results. To verify the feasibility of the system, field tests were conducted at 120-124 GHz frequencies and 4 GHz bandwidths. The system has a maximum operating range of 50 m, a beamwidth of 4 degrees, and a theoretical range resolution of up to 0.5 mm. By integrating optimization algorithm, the terahertz radar system can realize high-precision dynamic monitoring of coal level in the bunker, and show excellent anti-interference ability and stability, showing great practical application value.

The existing terahertz scanning equipment relies heavily on persons’ experience to identify image object. Because the resolution of the terahertz image is not ideal, the work intensity of the security personnel is very high, easy to get distracted or tired, so that long-term accuracy cannot be guaranteed. Therefore, it is of great necessity to intelligentize terahertz security inspection equipment to reduce manual labor intensity by deep learning technique. In this paper, we develop an automatic detection method of hazardous objects based on improved YOLOv8 for a terahertz security inspection equipment. The method realizes the automatic detection of dangerous objects by the improved YOLOv8 model. Specifically, the method incorporates Context Aggregation Networks into the YOLOv8 model to enhance its capability of feature extraction. To adapt to the low resolution of terahertz images, the neck network of YOLOv8 is designed as BiFPN. Additionally, the original C2f residual module is replaced with the C3 module to reduce model parameters and complexity, decreasing computational demands and increasing detection speed. Finally, EIoU is set as the target for model optimization. The experimental results show that the improved YOLOv8 model achieves a 95.2% mAP0.5 and 79.3% mAP0.5-0.95. The computational power requirement of the model is as low as 7 FLOPs and inference time is as fast as 1.4ms. With lower parameters and computations, the improved YOLOv8 model realizes improved detection accuracy and speed, outperforming current mainstream models including Sparse R-CNN, YOLOv5, and SSD, etc.

In the study of spatial target component recognition based on terahertz radar imaging using the traditional YOLOv5network, the recognition performance of the model decreases due to large overlapping areas of components in some samples and unclear imaging of small components. To address this issue, this paper proposes a typical component target recognition model, BoT-YOLO+, based on an improved YOLOv5 network architecture. On one hand, the proposed model enhances performance by introducing the BoTNet backbone architecture, which incorporates the attention mechanism from Transformers and improves the feature extraction capability for small components and thereby increasing the recognition rate of small components, without significantly increasing computational costs.

Cultural relics are a symbol of various times. The protection and restoration of cultural relics is a multidisciplinary work covering archaeology, physics, chemistry, and biology, etc. Exploring scientific means of nondestructive testing has always been the development direction of cultural relic protection workers. Terahertz has been widely used in various fields as an emerging discipline in the past 20 years. Utilizing the terahertz technology to the cultural relic has also been a research hotspot in recent years. This paper mainly describes the application of terahertz reflection imaging to a lacquer box. The experimental results show that the lacquer box sample has a double-layer structure. The bottom layer is a wooden basis, however, the upper surface is painted with lacquer and painted on it in order to achieve the purpose of decoration. The results demonstrate that the potential of terahertz reflection imaging can carry out non-destructive testing of samples with layer structure to study whether there is damage inside.

The conventional approaches to sub-terahertz metamaterial (MM) fabrication, which rely heavily on lithography and etching techniques, pose significant challenges in terms of costs, material consumption, and the complexity of the processes. To address these issues, there is an urgent need for enhancing efficiency and environmental sustainability in MM production. Thus, we explore a rapid customization approach utilizing ink-jet printing technology (IJPT) to achieve additive manufacturing of MMs for signal selectivity in 6G communication. The proposed MMs comprise two primary units: one double-split ring (DSR) hollowed MM (MM-1) is designed to realize band-pass filtering, while the other dumbbell-shaped MM (MM-2) has a band-stop filtering response. By adaptively adjusting the nozzle voltage between 14 and 22 V, we ensure a consistent line width of approximately 200 μm, enabling an effective frequency selection characteristic within the sub-terahertz band. Through altering the scale factor of MM-1, we achieve a frequency tuning from 61.00 to 72.75 GHz, with an optimal Q value of 27.71. Similarly, the periodic size of MM-2 results in a frequency shift from 76.58 to 91.10 GHz, with a maximum Q value of 11.26. Measured and simulated results for the MMs are in close agreement. The experimental results validate the feasibility of IJPT as an effective method for processing sub-terahertz MMs, emphasizing the great potential for optimizing6Gcommunication systems and signal filtering applications.

Since there is no atmospheric influence, space-based terahertz inverse synthetic aperture radar (THz-ISAR) can obtain high-resolution images of space targets at a certain distance, so it has important application prospects. In addition, the relative motion of space-based platforms and space targets is complicated. For some satellites that are tumbling out of control or are undergoing attitude change, the imaging angle formed by the target relative to the radar can be considered to be non-uniform, which in turn leads to imaging defocus.In this paper, a minimum entropy-based THz-ISAR rotation parameter estimation and motion compensation method for nonuniformly rotating space targets is proposed. Firstly, the radar echo is compensated for the translational motion and corrected for the range migration to obtain an envelope-aligned one-dimensional range profile. Then, the minimum entropy (ME) algorithm based on Newton’s method (NM) is used to estimate rotational velocity and centre of the targets, and to compensate the range second-order spatial-variant (SV) phase. After that, the golden section method is applied to estimate the target rotational acceleration and compensate the azimuth second-order spatial-variant phase. Finally, the cyclic variable method is used to iterate these two compensation steps alternately to avoid the coupling effect of range and azimuth second-order spatial-variant phase, while obtaining the target rotation parameters and the well-focused imaging results.

The design, fabrication and measurement process of silicon based waveguide transitions working from 500 GHz to 750 GHz is reported in this paper. Gap waveguide is applied to the waveguide cavity design. The electromagnetic bandgap characteristic helps to split the three dimensional structure into slices. Hence the waveguide transition could be composed of thin gold coating silicon layers. Micro-electro-mechanical system (MEMS) technology is used for the fabrication of periodic pins and multi-step transitions. A metallic clamp is designed for measurement. Measured results show that the transmission loss is better than 0.13 dB/mm and the reflection coefficient is lower than -9 dB.

With the rapid development of aerospace industry, multilayer structural composites are more widely used. Debonding, bubbles, inclusions and other defects in multilayer structural composites are in urgent need of new non-destructive testing methods for detection and identification. In this paper, the terahertz time domain spectroscopy system is used to detect the polymer materials with different embedded defects. In order to improve the accuracy of defect classification and recognition, the terahertz signal is transformed into the form of wavelet time-frequency graph through Morlet wavelet basis, which enrichesthe local feature and structure information. Then the ResNet101 network with residual block structure is used and compared with VGG16 and DenseNet201 networks. The experimental results show that Compared with the VGG16 network model, the ResNet101 network model has more stable training process, higher accuracy, and faster convergence speed. In training, the accuracy is raised to 100%, and the simultaneous loss rate is reduced to 0. Compared with the DenseNet201 network model, it has higher accuracy and more reasonable training process. It is proved that the ResNet101 network model is more suitable for the classification and identification of terahertz signals of defects in multilayer composite materials. This method can provide a new idea for the defect classification of multi-layer structure composites.

Millimeter wave radar system is featured by low power consumption, whose strong penetration, owing to the short wavelength all-weather operation. With the rapid development of modern detection, the need of high-quality fast imaging algorithm urgent for security detection. In this paper ,a three-dimensional imaging algorithm on frequency domain is proposed to visualize the target. In detail, a linear frequency modulated continuous wave (LFMCW) radar with working frequency from 60GHz to 64GHz is used to acquire data on which the target image is reconstructed. The basic principles of linear frequency modulation continuous wave radar are introduced in the beginning, then, the echo signal model of the single input single output (SISO) imaging system and the procedures of imaging algorithm are deduced at length The scanning imaging experiment of the target is carried out to evaluate the proposed algorithm, ending up with fine contour of the target. Besides, the resolution of the actual image is obtained by analyzing the image contour. Finally, theoretical resolution is calculated on condition that center frequency equals 62GHz and aperture equals 20cm in both vertical and horizontal directions. The comparison suggests that the actual performance of radar is consistent with the expected one.

Terahertz frequency is high, and which is suitable for the realization of large signal bandwidth. At the same time, terahertz wave has benign anti-stealth and anti-interference ability. Terahertz radar can obtain the fine scattering characteristic information of the target, which is of important significance for target fine detection, target recognition and terminal guidance. However, because of the huge difference between the size of targets and the wavelength in the terahertz, the scattered near-field region of the target is up to tens of kilometers in distance. Therefore, In the terahertz band, most of the scattering targets are located in the near-field region. When the terahertz radar detects the target, the electromagnetic scattering should not be similar to the plane wave, and therefore the far-field radar scattering cross-sectional area theory is no longer applicable. It is absolutely vital to analyze the near-field electromagnetic scattering characteristics of the terahertz band. This report mainly carries out simulation analysis, and which calculation method is the Large element Physical Optics method. Firstly, the variation of the RCS of the standard target sphere and plate in the terahertz band with the distance is calculated and analyzed in the near-field. When it degenerates to the far-field, the calculation result satisfies the analytical solution of the standard target RCS, which verifies the correctness of the calculation method. Secondly, to analyze the RCS of the complex aircraft scale target in the terahertz band throughout 360°, and drawn the RCS distribution of the rectangular coordinate system and the polar coordinate system.

Non-volatile phase-change materials (PCMs) provide a promising material platform to design reconfigurable electro-optic devices. However, for PCM-based spatial light modulators, electro-optic switching of material phases by a voltage pulse is generally affected by the filamentation effect. An alternative way to induce phase transition and reversible modulation is by on-chip micro-heaters, whereas these microheaters can only reflect light where noble metals are generally used as thermal resists. Moreover, the corresponding power consumption is very high because most heat energy dissipates into substrates, preventing the compact integration of PCM-based micro-modulators into the miniaturized and reconfigurable optical chip-array systems. Here, we propose a PCM electro-optical modulator working on the transmission mode with MEMS microheaters. The heating circuit is fabricated on a 200-nm-thick free-standing Si3N4 film, and transparent conductive oxides are used as thermal resists to allow light transmission. According to multi-physics simulations, the proposed MEMS modulators can make a thermal hotspot up to 400 °C in just 100 μs at 2 V bias voltage, indicating low power consumption. Meanwhile, by changing the phase state of PCM from amorphous to crystalline, the mid-infrared light transmission can switch from 0.71 to near zero in the mid-infrared range. The proposed MEMS modulators could inspire new applications in reconfigurable phase-change nanophotonics.

This paper presents a novel comprehensive least squares calibration method based on digital sideband separation receivers. Addressing the issue encountered in existing traditional calibration methods, which necessitate measurements of both amplitude and phase at intermediate frequencies (IF), the proposed comprehensive least squares calibration method requires only the measurement of the amplitude and phase of the digital backend output for calibration. During digital signal processing of analog IF sampled outputs at the digital backend, a calibration coefficient matrix generated by the algorithm is introduced, and the amplitude and phase of the digital backend output signals are recorded. Amplitude and phase imbalances are solved using the least squares method, while the condition number of the calibration coefficient matrix is employed to determine whether preconditioning of ill-conditioned matrices is necessary. This process yields the amplitude and phase imbalances of the In-phase (I) and Quadrature-phase (Q) channels, which are utilized to compute compensation coefficients for calibration. Through MATLAB simulation, the feasibility of the comprehensive least squares algorithm is verified. Results demonstrate that sideband rejection ratios (SRR) of over 50 dB are achieved when amplitude imbalances range from -2.21 dB to 3.62 dB and phase imbalances range from 5° to 70°. This approach provides a promising pathway to enhance the operational performance of digital sideband separation receivers in practical applications.

In order to solve the problem of mismatch between the size of the near-field probe and the wavelength of the terahertz beam in the terahertz near field imaging system, the quasi optical transmission path of the near field imaging is designed in this paper. The quasi optical path adopts two off axis parabolic mirrors for the collimation transmission and focusing of terahertz waves respectively. The electromagnetic simulation of the designed terahertz collimating transmission optical path is carried out, and the simulation results show that the collimating transmission and focusing of terahertz waves can be realized effectively. At the same time, the radiation performance of the near field probe in the electrically large and complex environment is simulated. The simulation results show that the quasi optical path can effectively collect the terahertz waves scattered by the probe. Based on the designed terahertz quasi optical transmission path, a terahertz near field imaging system is built and tested. The experimental results show that the quasi optical path can be effectively applied for terahertz near field imaging.

To address the complexity of full-array antenna simulation models and reduce computational demands, this study introduces an innovative design and simulation method for phased array antennas based on metasurfaces with composite periodic structures. This approach harnesses the arithmetic phase differences between array elements to engineer a phased array capable of beam scanning, thereby simplifying the complex initial model into a streamlined basic unit cell model with periodic boundary conditions, which significantly enhances the efficiency and speed of full-array antenna design and analysis. In this research, we constructed two types of phased arrays using microstrip patch antennas with composite periodic structures: a comprehensive full-array model and a periodic model. Finite element simulation results have confirmed a high level of consistency in antenna gain between these two models. The simulation and analysis method presented in this paper not only effectively reduces the complexity of the simulation model but also conserves computational resources without sacrificing the accuracy of the results, providing an effective computational strategy for the design and simulation analysis of complex full-array antennas.

Terahertz technology is the only technology that can achieve specific identification of hazardous chemicals, drugs, explosives, and other contraband inside mail packages without opening them. Therefore, overcoming the challenges of terahertz postal security inspection technology has broad market application prospects and extremely important social significance. In recent years, the combination of terahertz time-domain spectroscopy and deep learning has been widely applied in the field of material identification. However, in practical applications, based on the characteristic absorption spectra in the frequency range, the terahertz absorption spectra of amino acids vary with different packaging materials. Substance classification algorithms based on deep learning and machine learning show high accuracy in offline data models but lower accuracy during real-time online detection. Real-time detection of online amino acid samples based on terahertz time-domain spectroscopy technology should fundamentally solve these issues by increasing the training data, i.e., generating more data from the raw data. Generative adversarial networks (GANs) are a type of deep learning model that can learn the complex distribution of raw data. However, in the field of terahertz material identification, GANs have rarely been used to generate data to improve classifier performance. Therefore, this paper proposes a data augmentation method based on GANs. Then, a terahertz spectrum classification technique combining decision tree (DT), support vector machine (SVM), and convolutional neural network (CNN) is used to identify terahertz spectra within packages.

Astragalus is commonly used Chinese medicine materials, which has the effect of invigorating qi. Due to the different origin, the quality and efficacy of astragalus are different, which has an important impact on clinical application. This paper presented a new identification method of astragalus origin based on terahertz time-domain spectroscopy (THz-TDS) combined with machine learning. Five astragalus from Xinjiang, Shanxi, Inner Mongolia, Gansu and Qinghai were taken as the research objects. The terahertz spectral characteristics of astragalus from different producing areas were described, and the discriminant classification models were established by combining the absorption coefficient spectra in the range of 0.2~1.3THz with random forest (RF), backpropagation neural network (BPNN) and convolutional neural network (CNN). The results showed that CNN had the best prediction effect on astragalus origin, and the overall accuracy reached 95.2381%, indicating that terahertz time-domain spectroscopy combined with convolutional neural network model could be effectively used for automatic identification of astragalus origin, providing a new method reference for identification of Chinese herbal medicine origin.

The terahertz spatial light modulator (THz SLM) is a crucial device for the dynamic modulation of terahertz waves. It is beneficial in the fields of THz imaging and photonic computing. For example, we could encode and perform compressive sensing using THz SLMs. Compared with THz SLMs working in the reflective mode, the transmission mode THz SLMs (T-THz SLMs) show an obvious advantage in insertion loss. Due to the ultra-low insertion loss, several T-SLMs could be used in series, greatly facilitating its application of optical computing. In this study, we designed an electrically driven T-THz SLM, which the MEMS approach could fabricate. To enhance the performance of the T-THz SLM, the T-THz SLM was designed to be prepared for freestanding ultra-thin silicon nitride windows. Owing to the merits of ultra-low thermal capacitance and high transmission of the devices, a high-speed, low-power-consumption, low-insertion-loss, broadband T-THz SLM would be realized.