In order to build the modern integrated transport system and support the development of China's high-speed railways towards intelligence, based on the fifth-generation mobile communication technology(5G), the 5G for railways(5G-R) system will become the preferred choice for intelligent rail connectivity. The 2.1 GHz band is expected to host the 5G-R system to support new services that are emerging in the railway industry. The high-speed railway hub scenario with a large volume of complex types of services is the center for train operation command and the dispatching of passengers and the cargo. Thus, it places higher demands on the railway communication system. In this paper, the railway hub of Xiamen North Station is taken as the research scenario. A 3D model is built based on the electronic map. The Ray-Tracing(RT) method is adopted to simulate and analyze the channel and propagation characteristics of the scenario. Relevant suggestions are proposed for the design of the communication system. The results of this paper will help to provide a reference for the design and optimization of 5G-R systems for railway hub scenarios, and improve the quality of wireless coverage in railway hub scenarios.

High-speed maglev trains will be very competitive ground transportation tools in the future because of their high speed and convenience. However, high-speed maglev train-ground communication faces the Doppler effect caused by high-speed motion, frequent handovers, high real-time transmission requirements, and the requirements of carrying multiple services. In this paper, the adaptability of common civil communication and railway subway communication technologies such as Wireless Local Area Network(WLAN), 38 GHz millimeter wave and Long Term Evolution-Metro(LTE-M)/ 5G in high-speed maglev system is studied, and the analysis on the application of these communication technologies in the train-ground communication environment of high-speed maglev system is carried out. The direction and technical route of how to improve and apply these communication technologies are put forward.

The throughput of high-speed railway train-ground mobile communication system is not only limited by spectrum, but also limited by high mobility, which has become an important bottleneck restricting the development of intelligent high-speed railway. An effective way to improve the system throughput is to use millimeter wave band with wide continuous spectrum and large-scale Multiple-Input Multiple-Output(MIMO) and multi-beamforming technology. However, the multi-beam streams with fixed beam width will result in serious inter-beam interference when the train is running at high speed. An adaptive beamforming and joint transmission scheme based on optimal beam width is proposed in this paper. When the train is close to the Base Station(BS), all beams are activated to realize spatial multiplexing thereafter to improve system capacity and transmission reliability. When the train is far away from the BS, in order to avoid serious interference between beams, more antenna array elements are utilized to form fewer beams to obtain greater beam gain, and beams of the adjacent BS are cooperated to transmit to improve the throughput of the system. Simulation results show that the performance of the proposed scheme is significantly better than that of the existing beamforming and traditional adaptive selection schemes.

In order to promote the overall construction of the intelligent railway and to realize the intelligent high-speed railway, the communication infrastructure of high-speed railway will apply the 5th Generation Mobile Communication Technology(5G) to build the 5G for railway system. 2.1 GHz band is expected to be the bearer band for the 5G for railway system. The high-speed rail platform is the functional core of the loading and discharging of passengers and the cargo, and its propagation characteristics at 2.1 GHz need to be studied urgently. The wireless channel in the high-speed rail platform of Xiamen North Station at 2.1 GHz is characterized through Ray-Tracing(RT) simulations. Based on the simulation results, the channel parameters in the high-speed rail platform scenario are analyzed, including Root-Mean-Square Delay Spread(RMS DS), azimuth angular spread of arrival/departure, elevation angular spread of arrival/ departure(ASA/ASD, ESA/ESD), Cross-Polarization Ratio(XPR) and cross-correlations of the above parameters. Based on these results, relevant suggestions are provided for the design and evaluation of the 5G for railway system in the high-speed rail platform scenario.

According to the characteristics of dense layout and high power consumption of the 5th Generation mobile communication technology-Railway(5G-R) base stations, combined with the requirements of railway wireless communication system, the Photo-Voltaic(PV) power supply application scheme of 5G-R Remote Radio Unit(RRU) equipment is studied. By comparing and analyzing the daily power consumption of 5G-R RRU equipment and the daily power generation of solar cells, the power supply system using PV power combined with storage and external power supply is determined. By comparing the economy and reliability of several combined power supply structures of PV power and external power supply, the power supply structure switching on the Direct Current(DC) side is recommended, and the PV power supply application using DC side intelligent power distribution is proposed. In this study, while ensuring the safe and reliable operation of 5G-R system, the energy saving of railway 5G-R system is realized by using PV power supply.

In October 2018, IEEE released the latest WiFi standard protocol 802.11ax(WiFi6). It can support both 2.4 GHz and 5 GHz frequency bands with the maximum transmission rate of 9.6 Gbit/s. At present, WiFi6 has been applied in smart home, radio, and mine construction. However, it is scarcely applied in the field of rail transit. An integrated urban train ground communication system based on WiFi6 is proposed. Aiming to the serious competition for frequency resources between production businesses, the first actual subway line test of the WiFi6 system in the country is conducted on the Dalian Metro Jinpu Line. The entire test process completely replicates the actual running scene of the train, including real vehicles, equipment, and actual communication scenarios such as elevated and tunnels. Extensive test results show that the designed WiFi6 system bears the characteristics of large network throughput, fast communication and transmission rate, long battery life, and strong AP signal coverage, satisfying the rail transit business. It can guarantee the reliable transmission of Communication-Based Train Control(CBTC) traffic and provide an efficient transmission path for Closed Circuit Television(CCTV) and Passenger Information System(PIS) traffic.

With the explosive growth of wireless data rate, the terahertz band(0.1~10 THz) has attracted increasing attention due to its rich frequency resources. The terahertz photonic communication technology has advantages such as broadband, high modulation efficiency and low harmonic distortions. It is expected to greatly promote the wireless transmission data rate toward the Tbit/s level. Based on the review of the key technologies for realizing ultra-high-rate terahertz photonic communication, including terahertz communication transceivers, digital baseband signal processing technologies, terahertz photonic communication system architectures and experimental demonstrations, the future development trends of the terahertz photonic communication technologies are discussed. The application scenarios of terahertz photonic technology from the micro to macro scales are also analyzed.

Recently, Terahertz Time Domain Spectroscopy(THz-TDS) has been proposed to measure electron density of plasma. Nevertheless, the method still needs to be further verified by estimating the reliability and practicability using theoretic or other experimental methods. In this work, the THz-TDS technique is applied to measure the electron density of Inductively Coupled Plasma(ICP). The ICP discharge is operated in the H-mode with a RF power supply of 13.56 MHz at various input powers from 250 W to 400 W under argon pressures from 39 Pa to 99 Pa. It is observed that electron densities monotonically increase with the increase of RF input power and argon pressures, and the measured electron densities are in the range of 1013~1014 cm-3 under the given conditions. Meanwhile, a fluid numerical simulation based on COMSOL Multiphysics using drift-diffusion model is adopted to investigate the characteristics of the discharge plasma at the given powers and pressures. The results show that the simulated electron densities are in qualitative agreement with the measured results, which supports that the measurement method using THz-TDS is feasible. Furthermore, the variation trends of the electron density with the pressures obtained by simulation and measurement are also similar, which further verifies the application potential of the THz-TDS for non-invasive plasma electron density measurement.

Aiming at the problem that the image quality fluctuation of passive terahertz security system is caused by environmental change, which affects the recognition algorithm and leads to a significant decrease in accuracy, this paper proposes an improved YOLOv4 algorithm based on Focal-Efficient Intersection Over Union(EIOU) loss function, and uses passive terahertz human security image to conduct model training for prohibited items of knife and gun. A terahertz image database of people carrying suspected objects in different environments and different positions is established, and a rich data set is constructed by image augmentation method. The Complete IOU(CIOU) loss of YOLOv4 is improved to Focal-EIOU loss to improve the robustness of the algorithm for terahertz image recognition, and then a better model is obtained after training. In the test set of this paper, since YOLOv4 algorithm has low robustness for terahertz image recognition accuracy, CIOU loss of YOLOv4 is modified and adjusted to Focal-EIOU loss, and a better model is finally obtained through training. The mean Average Precision (mAP) of the model trained by the improved algorithm reaches 96.4%, the detection speed is about 28 ms, and the average value of IOU is 0.95, which are higher than those of the conventional algorithms under the same conditions, the detection and recognition effect are improved. The experimental results show that the proposed method can effectively improve the suspect identification accuracy of passive terahertz human security system, which is conducive to the popularization and application of this technology in the field of human security.

There are two typical difficulties in radar detection of uncooperative targets: weak target and unknown velocity. Coherent pulse integration method is used in detecting the weak target, but it needs to detect the target velocity accurately and compensate the velocity, therefore it is limited in practical application. A design method of coded stepped frequency chirp waveform is proposed, which can integrate the pulse when the velocity is unknown and synthesize the wideband after the velocity is measured accurately. Firstly, the range velocity coupling characteristic of Linear Frequency Modulation (LFM) is studied, and the range migration is compensated by waveform parameter design. Then, the wideband synthetic technology of the stepped frequency chirp waveform is studied in order to get high resolution range profile without range ambiguity. Finally, simulation experiments validate the advantages and practical value of the proposed waveform.

In order to study the echo mechanism of Mesosphere-Stratosphere-Troposphere(MST) radar at tropospheric height, the phenomenon of abnormal echo power in the upper troposphere is studied by using the radar echo power and wind field under special weather conditions(cut off low pressure), Lomb-Scargle spectrum analysis, harmonic fitting, Richardson number and other methods. Experimental and simulation results show that under the cutting off of low-pressure weather conditions, persistent gravitational waves are generated by the jet stream and deep convection, and the gravity waves are disseminated from the bottom up beneath tropopause and saturated broken, which dissipates energy in the atmosphere, destroys atmospheric static stability, and forms the Kelvin-Helmholtz(K-H) instability layer, and then develops into turbulence, thus making the region of radar echo power increased.

The depth of traditional Covariance Matrix Taper(CMT) null widening algorithms will become shallower after the null broadening under high dynamic conditions. A novel null widening algorithm based on Laplace distribution is proposed. Firstly, the signal model is reformulated and the two-dimensional Laplace algorithm can be derived. Secondly, the algorithm extracts the interference components by projection transformation from the data, and enhances the interference components by weighting the coefficients. Then, null widening algorithm is adopted to expand interference direction based on the reconstructed covariance matrix. Finally, the Power Inversion(PI) algorithm is utilized to suppress the interference. Simulation shows that the depth of the proposed algorithm is increased on the basis of the null widening algorithm, and the robustness of the anti-jamming algorithm is improved. Being applicable to arbitrary planar arrays, the algorithm possesses great practical significance in engineering.

Since deep learning algorithms have outstanding advantages such as strong feature expression ability, automatic feature extraction, and end-to-end learning, more and more researchers have applied them to the field of communication signal recognition. However, the discovery of adversarial examples exposes deep learning models to potential risk factors to a great extent, which has a serious impact on current modulation recognition tasks. From the perspective of an attacker, adversarial examples are added to the currently transmitted communication signal to verify and evaluate the attack performance of the target countermeasure sample to the modulation recognition model. Experimental results show that the current targeted attack can effectively reduce the accuracy of model recognition, and the constructed logit indicator can be better applied to measure the targeted effect more fine-grained.

An Ultra-Wide-Band(UWB) antenna with flexible material Polydimethylsiloxane(PDMS) as the dielectric substrate is designed. Its radiating element is composed of rectangular and semi-circular metal patches. The two-part structure is connected by a strip patch. The coplanar waveguide feeding method is adopted for the antenna. The above-mentioned metal patch is embedded inside the middle of the PDMS substrates; and this antenna can be used in a wearable system for human body communication. The embedded structure can effectively improve the flexibility and durability of the antenna. In order to better simulate the working environment of the antenna on human body, a three-layer human body tissue model including skin, fat and muscle is constructed. The maximum antenna specific absorption rate is 16.4mW that meets the international standards for human safety. Finally, the antenna samples are tested, and the results show that the overall performance of the antenna is stable, and the impedance bandwidth is 5.39-10.33 GHz, which meets the requirements of medical band(5.725-5.875 GHz).

Aiming at the non-smooth characteristics of the non-smooth devices/circuits, a modeling and simulation method based on the standard mathematical programming problem is proposed. Firstly, the non-smooth device is modeled as a Piecewise Linear(PWL) function with a possible state jumping. Secondly, some dynamic system equations of the non-smooth linear characteristic of the circuit formed by the non-smooth devices are given. Then, the time discretization is applied to these non-smooth dynamic system equations, and various types of one-step non-smooth problems are yielded, such as (linear) complementary problems or nonlinear(or quadratic) programming with equation-inequality constraints. The numerical solution is carried out. The simulation results show that the proposed numerical modeling method is effective for non-smooth systems with a large number of events and is robust to the variation of model parameters.

In order to measure the bolt preload with acoustoelasticity accurately under the low voltage excitation, a signal processing method based on correlation operation is utilized to process the echo signal. In this method, the ultrasonic propagation time is obtained by comparing the most similar position of the transmitted wave in the echo, so as to measure the preload. The correlation operation can suppress the noise, which can realize the measurement of transmission time under the condition of low SNR. At the same time, the method combined with interpolation can improve the resolution of time measurement. The Linear Frequency Modulation(LFM) pulse is chosen as the emission wave to test the preload of the bolt using acoustoelasticity under different voltage excitation. The maximum error is below 2.9%. The test results show that the method can realize the accurate measurement of ultrasonic propagation time under the condition of low voltage excitation, and the resolution of time measurement can be improved by combining the interpolation method.

A low-light-level camera system is designed based on the scientific Complementary Metal Oxide Semiconductor(CMOS) image sensor LTN4625 in order to meet the requirements of the high resolution and high dynamic range low light imaging system. According to the readout structure characteristics of pixel of LTN4625,the FPGA is taken as the control core of the camera system and an image fusion algorithm based on tensor optimization is proposed. The dynamic range of the camera is extended effectively and the image quality is improved. The imaging test results show that the camera can output pictures at 50 fps and the resolution of image is 4 608×2 592. The dynamic range of image is improved to 54.7 dB from 5.2 dB for low gain data and 11.4 dB for high gain data. This design is a kind of low-light-level camera design method with high dynamic response range and good imaging effect under weak light environment.