Multiband terahertz metamaterial absorbers are important photonic components responding to, manipulating and modulating terahertz waves. A multiband terahertz metamaterial absorber is constructed based on a periodically split ring resonator structure. Simulations and experimental tests show that the metamaterial absorber exhibits near-perfect absorption of incident terahertz waves at 0.918 THz and 1.581 THz in the Transverse Magnetic(TM) polarization case. Based on the dielectric sensitivity of the resonant absorption peaks of the device, the sensing performance of the metamaterial absorber is investigated after loading four kinds of pesticide solution with different concentrations of carbendazim, tricyclozole, paraquat, and thidiazuron, and the detection sensitivities of the device for the four kinds of pesticides are obtained as 1.06 GHz/ppm, 0.65 GHz/ppm, 0.67 GHz/ppm, and 2.07 GHz/ppm, respectively. The results show that the device can achieve the sensing detection of trace pesticides, which provides a new idea for future food quality and safety control.

As an artificial composite material with extraordinary physical properties, metamaterials can break through the limitations of conventional materials and open up a new way to design advanced functional materials. Terahertz waves have the characteristics of low photon energy, no ionization damage to biological substances, and molecular fingerprints etc. By combining with metamaterials, high-sensitivity detection of biological substances can be achieved, and it has attracted more and more attention from scholars at home and abroad. This article summarizes the progress made by terahertz metamaterial sensors in the field of biomolecule and cell detection in recent years. Firstly, the sensing principle and performance indices of the terahertz metamaterial sensors are introduced. Secondly, the development of the terahertz metamaterial sensors in the field of biological detection is described from the aspects of metamaterial structure design, substrate selection, and combination with microfluidics and new materials. The sensitivity of the metamaterial sensors can be further improved by optimizing the structure of the metamaterial, using a thin substrate with low dielectric constant, combining microfluidic technology or attaching a new material coating to the sensor, and their functions in biomedical detection can be enriched. Finally, the development trend and prospect of terahertz metamaterial sensors are envisioned.

Affected by external environment, flight path of airborne Synthetic Aperture Radar(SAR) is always inaccurate. Under the circumstance of limited accuracy of navigation system, estimation of Doppler centroid frequency from airborne SAR data is necessary. Error of Doppler centroid frequency estimation decides the linear phase error, affects the estimation of high order phase error and azimuth compression quality. In airborne terahertz video-SAR system, submillimeter motion compensation accuracy is required, which tremendously increases the importance of Doppler centroid frequency estimation. Traditional Doppler center frequency estimation methods usually have good results in the frontal view or small squint mode of SAR system, but the deviation is often large in the mode of high squint angle. In order to effectively complete range walk correction in multiple modes, based on the measured data of THz video-SAR and starting from traditional envelope estimation method, an improved envelope estimation method is promoted. Compared with two traditional methods, in front view mode, all methods mentioned can give good results, but in scan SAR mode, 97% images can be well-estimated and focused with promoted method better than with other methods. By comparison, the proposed method exhibits higher robustness and efficiency. This work can help research on SAR imaging under high frequency.

The trajectory tracking control of Underactuated Surface Vehicles(USV) with non-diagonal inertia matrix and damping matrix is studied. The global asymptotic stability of the closed-loop system is achieved by using the controller based on generalized dynamic inverse and sliding mode variable structure method. At the same time, the system singularity caused by dynamic generalized inversion is overcome by introducing dynamic scale inverse. Since zero projection matrix is not invertible, disturbing zero projection matrix is applied. Finally, the effectiveness of this method is proved by Matlab/Simulink simulation.

In order to detect and separate the information of two Radio Frequency Identification (RFID) collision tags, a detection method based on inherent memory characteristics of FM0 tag coding in Gen2 standard is proposed. By analyzing the characteristics of FM0 bit coding and memorylessly detecting the collided tags information, the conditional error probability of memoryless detection method based on the duration of single bit and the bit error rate of single tag information detection are obtained. Then a pair of measurement results corresponding to the first bit and another pair of measurement results corresponding to the next bit are got by using the characteristic that encoding of a single FM0 bit requires memory of the previous bit. The conditional error probability and bit error rate performance for one bit memory auxiliary detection of collided tags information are obtained thereafter. The performance of total delay time reduction for N tag groups is also analyzed in framed Aloha medium access scheme where the proposed detection method is adopted. The simulation results show that the proposed one bit memory auxiliary detection method has better Bit Error Rate(BER) performance than memoryless detection, and it can also reduce the total delay time of tag group accessing.

The key factor restricting the maneuverability of mobile radar is the system electrical grounding. By the theoretical research and actual verification of vertical grounding grid, horizontal grounding grid, and complex grounding grid, the approximate calculation method is proposed for electrical grounding resistance of different grounding grids. On the basis of theoretical derivation, the estimated algorithm for the mobile grounding network is proposed. The proposed algorithm is proved to be simple and feasible by the measurement results. The method has been applied to a certain type of mobile radar and provides important technical support for the fast deployment task of mobile radar.

The real-time, deterministic and reliability requirements of time-sensitive services in spatial information networks pose a huge challenge to traditional ethernet switching based on store-and-forward. An on-board time-sensitive network scheduling algorithm based on gate control is proposed, and a spatial deterministic ethernet switching scheme is designed. Aiming at the characteristics of satellite service burst, a gate control list generation method based on time division multiplexing is proposed, and the deterministic delay and delay jitter performance of the scheduling algorithm in the context of different traffic flows are simulated and analyzed. The simulation results show that the minimum delay of the proposed algorithm is reduced to the order of 10 μs, and the minimum delay jitter is 0, which meets the deterministic transmission requirements of space time-sensitive services and improves the deterministic, reliability, and flexibility of spatial information network service transmission.

The identification of enemy air target's combat intention is an important part of battlefield situation awareness. In order to improve the accuracy of Intention Identification by making full use of the temporal correlation of the detected air target motion state information, an identification of enemy air target's combat intention method is proposed based on Long Short Term Memory(LSTM) network in this paper. Firstly, the simulation platform is utilized to deduce large number of air enemy target data according to four different intentions. Then, the generated data is cleaned and slide window processed to obtain the effective sample set. Finally, the LSTM network is adopted to learn the generated sample set and form the combat intention identification model of enemy air targets. Experimental results show that the prediction accuracy of network can reach 92% by using the LSTM network to learn the time-related characteristic information of data, and the result is better than that of traditional classifier.

In order to form a set of measurement system dedicated to the closed resonant cavity to replace the vector network analyzer, the spectrum analyzer module and the tracking generator module are adopted to control the system based on the C++ and VISA library functions and achieve the measurement function on the resonant frequency and the quality factor. Thereafter, the dielectric constant of the microwave dielectric material can be calculated, and a set of integrated closed cavity dielectric constant measuring system is formed. The system is employed to measure the microwave dielectric ceramic material K65 compared with the vector network analyzer. The relative error of the dielectric constant is 5.5×10-3, the relative error of tanδ is -3.74×10-2. When measuring material K35, the relative error of the dielectric constant is -1.69×10-3, and the relative error of tanδ is 1.08×10-1. The relative error of the dielectric constant is less than 0.01, and the relative error of tanδ is less than 0.5, indicating that the measurement results of the integrated dielectric constant measurement system are accurate. It can be used for the dielectric constant measurement by the closed cavity method, and can also be extended to other dielectric constant measurement systems.

Millimeter wave communication is capable of supporting multi-gigabit wireless access thanks to its abundant spectrum resource. Nevertheless, the severe path loss makes it vulnerable to blockage events especially in dense urban environments. To address this issue, Intelligent Reflecting Surface(IRS) as a new technology has been introduced to provide effective reflected paths to enhance coverage of millimeter wave signals. This paper mainly discusses the millimeter wave systems assisted by IRS, deduces the average accessible rate of IRS, and compares the performance of this technology with the traditional Amplified and Forward(AF) relaying system. The analysis shows that when the number of reflecting elements reaches a certain value, the average accessible rate of IRS system is higher than that of AF relaying system. The optimal value of the reflecting elements is obtained to maximize the energy efficiency of the IRS system. The simulation verifies the correctness of the derived formula and shows that the energy efficiency of the system increases with the increase of the reachable rate when the number of reflecting elements is at the optimal value.

Abstract：In view of the "ghost" problem caused by the interference of foreground pixels in the model initialization of Visual Background Extractor(ViBe) algorithm, and facing the problem of updating strategies for complex contextual environments, the initial optimization of the ViBe algorithm model using the image pixel mean as a reference is proposed. At the same time, an adaptive updating strategy varied with the complexity of the background model is proposed. Firstly, the background model is initialized by using the similarity between neighborhood pixels and continuous frame background pixels. Then whether the background model is stable is determined by calculating the variance of each pixel between the samples, in order to establish the adaptive update strategy. Lastly, the moving target is extracted. The validation of CDnet2014 data set shows that this algorithm can effectively improve the phenomenon of"ghost". It not only improves the robustness of the background model in the complex environment, but also improves the objective evaluation indexes.

To improve the reconstruction quality of hyperspectral compression sensing, a hyperspectral reconstruction algorithm based on dual-band prediction is proposed. K-means clustering algorithm is introduced to adaptively complete the band grouping, and the dual reference bands in the group are determined. Dual band prediction model is employed to obtain the initial prediction image. Based on the predicted image, the modified reconstruction and weighted fusion method are utilized to achieve highprecision image reconstruction. The results show that at the same sampling rate, the reconstructed image of this method has better Peak Signal to Noise Ratio(PSNR) and Structural SIMilarity(SSIM) compared with that of the existing reconstruction method.

The defects of large randomness and few actual samples in the detection of malicious domain names would lead to the overfitting in deep learning model training. A malicious domain name detection method based on group convolutional neural network is proposed. Firstly, the domain name is converted into embedded word vector representation; secondly, a random data set is generated through a combination of random dimensions and convolutional neural network groups are constructed. The Inception structure is added to the network due to its advantages. For the imbalance problem of the inter? class samples, the inter?class balance coefficient is introduced to suppress the model training overfitting and improve the model generalization ability. The experimental results show that the constructed model can effectively detect malicious domain names on the collected domain name detection data set; after parameter optimization, the group convolutional neural network improves the detection accuracy of the constructed domain name detection set by 4% and 1% respectively compared with the shallow model combination classifier and the typical deep neural network model Long Short-Term Memory Convolutional Neural Network(LSTM?CNN), which reaches 98.9%.

Aiming to the problem of video target tracking in complex environment, a hybrid visual tracking method based on similarity optimization is proposed. Firstly, local cosine similarity is utilized to measure the similarity between target and candidate template, which can effectively suppress impulse noise caused by occlusion and light mutation, and improve the template matching accuracy. Secondly, discrimination weights of local targets are deduced based on the quadratic programming method of objective function, which effectively improves the discrimination ability of algorithm to target and background. Finally, in the process of system updating, discriminant updating of template is introduced, which effectively improves the model drift problem. The experimental results show that the proposed method can improve the tracking robustness and accuracy in complex and challenging environments.

In order to ensure effective monitoring of Wireless Sensor Network(WSN) coverage and connectivity maximization and energy consumption minimization, a sensor node deployment algorithm based on Multi-Objective Biological Habit Incentives(MOBHI) is proposed. Firstly, the area(territory) of the sensor nodes is marked according to multiple objects such as maximum coverage, maximum connectivity, and minimum energy consumption based on the scent marking behavior of the territory predator, and its monitored location is identified by imitating the scent matching. Secondly, multiobjective optimization problem is decomposed into multiple single-objective optimization subproblems and optimized simultaneously by applying the non-dominated Pareto optimization, so as to obtain the solution of the desired objects. The simulation results show that the proposed algorithm is superior to other multi-objective and single-objective optimization algorithms in sensor nodes deployment in terms of network coverage, connectivity and energy consumption.

The encapsulation principle, design method and measurement results of a miniaturized comparator is introduced in this paper. The difficulties of miniaturization and measurement are tackled. A bridge chip is attached on the surface of the Low Temperature Co-fired Ceramic(LTCC) substrate, and the wire is led out from the inside of the substrate to form a test port on the surface of the substrate. Design scheme of miniaturization and high integration is realized based on micro assemble technology. The package shell is designed to realize the airtight package of the comparator and improve the reliability of engineering application. According to the measurement results of the module, the standing wave ratio is less than 1.8, the amplitude consistency is better than ±0.8 dB, the phase consistency is less than 10°, and the mass is 35 g.

Capacitance voltage dividers are often utilized to monitor voltage values in high power electrical pulse devices. The amplitude linearity of the capacitive voltage divider characterizes the consistency of the monitored voltage value. In this paper, the principle of the dielectric substrate capacitance divider is analyzed, and the performance of the dielectric substrate capacitance divider is simulated under the rectangular pulse high voltage source(500 V~4 kV) with a pulse width of 200 ns and the impact high voltage source(10~100 kV) with a half-width of 10 μs respectively. The design and fabrication are carried out. The results show that the amplitude linearity of the ceramic capacitor voltage divider is good, which is 1.16% within the range of 500 V~100 kV. The voltage divider calibrated under small signals can be applied to the measurement of large signals.