Abstract： The orthogonality of different orbital angular momentum (OAM) modes of vortex electromagnetic wave can be used as a new dimension of information muti-use freedom to further improve the communication capacity. A vortex electromagnetic antenna system based on double uniform circular array is designed, which can generate ± 1 and ± 2 four mode vortex electromagnetic signals at the same time. The results of anechoic chamber test and system test show that the designed antenna system has well isolation between different modes, completes the data transmission of four BPSK signals without error code, and achieves significant improvement of system communication capacity.

Abstract：This paper introduces an engineering realization of an S-band 500 W high phase-stable solid-state power amplifier.According to the engineering requirement,four power chips are used for high powersynthesis,and the solid-state power amplifier with output power＞600 W is realized in the frequency range of 2 025 MHz~2 120 MHz.Under the environment temperature condition of 0℃~30℃，the phase change of 1 W~500 W output power is ＜11.5° by the low additional phase shift design,modular design of diffirent microwave plate units,power back-off and other measures,which achieved the engineering and technical requirement of the centimeter-level TT&C for S-band solid state power amplifier.

Abstract： This paper proposed a novel 4 in 1 high power airtight amplifier which worked at watt level for 3 mm band power combination use. This design introduced a silicon based structure to achieve airtightness for W-band components, solving the problem of traditional 3 mm band waveguide components being difficult to be airtight. It was designed by using the airtight waveguide structure. The software was also used for simulating the waveguide combiner. Finally, the power amplifier was carefully fabricated and tested. Results show that the final indicators meet design requirements.

Abstract： In this paper, a rectangular waveguide band-pass filter with a center frequency of 140 GHz is realized by using modern Computer Numerical Control (CNC) metal milling technology. The proposed filter is designed based on HFSS and then fabricated by CNC metal milling technology. The measured results agree well with the simulated ones indicating that the filter has excellent performances: the center frequency is 140.2 GHz, the relative bandwidth is 10.1% at 3 dB, the insertion loss is less than 0.5 dB, the in-band return loss is better than 25 dB, and the out-of-band isolation is larger than 30 dB at ±20 GHz away from the center frequency. The results further confirm the feasibility of 140 GHz waveguide band filter fabricated by CNC metal milling technology.

Abstract： Terahertz wireless energy transmission has potential applications in space. This paper introduces the main methods of terahertz generation and reception at present. In view of the phenomenon that terahertz waves can be wirelessly transmitted between subwavelength lithium niobate wafers, the possibility of enhancing the evanescent field on the wafer surface using on-chip microstructures is analyzed, which can increase the distance of terahertz wireless energy transmission between the lithium niobate wafers, providing a feasible scheme for the application of terahertz wireless energy transmission.

Abstract：A miniaturized and reusable receiving front-end applying to satellite communication is designed in the paper, which converts RF signal of 950 MHz to 2 150 MHz to IF by preselected filtering, amplifying and mixing. The design scheme of the circuit is introduced first, and the key indicators are analyzed. The receiving front-end is 105 mm×55 mm×9.5 mm in size. Miniaturization and universal were realized in the paper, which can be reused in various satellite communication devices.

Abstract： In this paper, an Hammerstein-Wiener model based on extreme learning machine is built to identify Continuous Stirred Tank Reactor(CSTR) nonlinear system which is used in chemical process widely. In the proposed Hammerstein-Wiener model, the two nonlinear blocks are described by two different extreme learning machine neural networks. The linear block is described by ARX model. Due to the special structure of the extreme learning machine, this model can be expressed in the form of linear regression. The model parameter identification is achieved by generalized least square algorithm. The identification process is simple with less computation complexity. The simulation result shows that this proposed approach is effective. Compared with Hammerstein model based polynomial and ARX-LSSVM Hammerstein model，the proposed method has higher identification accuracy.

Abstract：Aiming at the problem of the high probability of miss and false positive rates in single-view Mammography, an improved automatic detection algorithm is proposed in this paper. The dilated residual network (DRN) combined with a modified feature pyramid network (FPN) is used for the detection of breast masses. The expansion convolution in DRN is used to reduce the number of downsampling of images. The number of layers of the DRN is also increased to satisfy the required input of the FPN. In the FPN structure, the attention mechanism is used to reduce the information loss caused by the direct fusion of different feature maps, while dense connections are used instead of the original lateral connections to make full use of the location and detailed information on the target for the shallow features. Simulation experiments show that the detection accuracy of the designed model on the CBSI-DDSM dataset is improved by 7.1 percent compared to the baseline.

Abstract： With the rapid increase of SoC size and complexity, how to implement full and complete verification at the system level has become an important challenge. Difficult work not only comes from the scheduling of various hardware resources at the system level and the implementation of test scenarios, but also comes from how to obtain useful information from complex system operations. In order to let the verifier master the system information in time during the simulation process, a SoC oriented omni-directional system monitoring verification scheme can obtain various system status information in real time based on the universal verification methodology (UVM) framework. The scheme is also based on an automated verification platform, and the monitored information is processed in this platform to help the verifier better analyze the system and locate problems.

Abstract：In order to solve the cache write back missing problem of the Spike verification model，this paper presents a superscalar cache model based on the reduced instruction set computer(RISC-V) instruction set。This cache model is based on the Spike environment, the cache model implements the basic principles of superscalar cache. This cache model implements the high level cache write back and write through process, and it is verified in the system level chip (SoC) simulation. This scheme can be used as a model for front-end logic verification, and it can complete large scale verification based on the RISC-V instruction set in a shorter time.

Abstract：Aiming at the requirements of miniaturization of satellite equipment and lightweight of the whole system, this paper proposes a design scheme of radiation hardened POL power supply based on top heat sink. This paper focuses on circuit design, structure design and process design. The rationality of the design is verified by simulation analysis, sample test and evaluation test in this paper. The sample designed based on this scheme has obvious advantages in volume, power density, weight and other aspects, which can provide a miniaturized and lightweight tertiary power supply solution for space applications requiring high reliable and high performance electronic equipment systems.

Abstract： In this paper, aiming at the research vacancy of dedicated communication engine in China, a design of configurable communication engine supporting multiple protocols is implemented. Taking the typical data link layer protocol, the engine block implementation of high-level data link control (HDLC) protocol, as an example, the simulation platform is built by System Verilog, and the test case is written in C language to ensure the correctness of the design by loop verification. The configurable engine block takes the self-developed RSIC core as the core, adopts AHB bus interconnection, integrates communication protocols such as HDLC and UART and general peripherals such as DMA, TDM and GPIO, etc., and realizes the processing of communication protocols and data transmission, which helps to free the processor load and improve the data processing efficiency. At the same time, the combination of HDLC and configurable communication engine solves the problem of high utilization rate of processor resources by HDLC with multi-channel signals.

Abstract：At the moment, an increasing number of high-voltage transmission lines must pass through substantially ice-covered locations, and transmission line breakdowns caused by snow and ice disasters occur on a regular basis, resulting in economic losses. As a result, it is required to investigate an effective detection technique for determining the state of ice-covered transmission lines in harsh conditions. This work presents and examines a weighted fusion approach for wavelet transform mode maximum and multi-scale and multi-structure morphology. Image preprocessing is performed on the acquired images using methods such as de-fogging, filtering, and gray augmentation, and the edge detection algorithm is achieved. By comparing the classical and commonly used edge detection operators, it is discovered that the error of this method is only 0.661 7% for the edge profile detection of ice-covered transmission lines under extreme conditions, which meets the requirements for the identification and detection of ice-covered transmission lines under extreme conditions. It provides technical assistance for calculating ice thickness and diagnosing transmission line faults.

Abstract： It is difficult to effectively balance the positioning accuracy, construction cost, and deployment flexibility of the current asset positioning system. To address this challenge,this paper designs an asset positioning system based on bluetooth low energy (BLE) mesh using multidimensional scaling-map (MDS-MAP) positioning algorithm. Firstly, the original received signal strength indicator (RSSI) value is filtered by Gauss-Kalman fusion to improve its accuracy. Then, time to live (TTL) is used to restrict the relay nodes, which improves the data transmission efficiency. Finally, in order to reduce the ranging error, the range matrix is modified by using the radius compensation method and the Bellman-Ford fusion iteration scheme. The experimental results show that the designed system can effectively update the bluetooth tag information and display the location, and the average positioning accuracy reaches 0.94 m. This system has the advantages of low cost and convenient engineering implementation, and has certain application value and development prospect.

Abstract：In order to solve the problems of multiple data types, large data volume and heterogeneity in radio regulation practice, how to manage radio monitoring data has been a hot topic of research work. Based on the radio regulatory knowledge graph framework, a radio monitoring data management graph model is established, and the signal modulation characteristics of typical radio services under Gaussian channel conditions are simulated with MATLAB as a tool to obtain a radio spectrum simulation dataset, so as to establish a knowledge base of radio spectrum usage conditions and realize the functions of knowledge query, reasoning and visualization. The experimental results show that the Django-based framework and the graph database Neo4j radio monitoring data management Web application system can provide an effective solution for the intelligent management of radio monitoring data.

Abstract：Electronic nose sensor array has the characteristics of high dimension and strong cross sensitivity. Improper sensor combination may lead to overlapping selection and poor performance. At present, array optimization algorithms based on information theory often rely on individual discriminate ability for feature selection. It is difficult to estimate the multicorrelations of features, and lack of consideration of feature complementarity and dependency. To solve these problems, a sensor array optimization algorithm based on fuzzy mutual information (FMI-SAO) is proposed, which estimates feature redundancy through fuzzy joint matrix and uses forward iterative selection strategy to mine features with high correlation, low redundancy, high complementarity and high dependency. Experiments show that the sensor array obtained by FMI-SAO can achieve higher accuracy in different pattern recognition algorithms.

Abstract： The intelligent computing is a critical factor in the SCADA software system. It has great advantages and convenience in processing logic relation between the tags, computing the cumulative value, the average value and the integral value of the tags relating to timing, and applying the steam meter functions in the thermal field. In this paper, a design way on intelligent computing in the control system is studied. It describes the concept on some nested calls between the combination of macros and the combination of tags firstly. Afterwards it introduces the processing flow of the computational module, the basic definition of the combination of macros, and the software module division. It describes the design ways on the computational configuration tool and the memory index table, on the global variables and the local variables, and on balanced load. Lastly it designs the program classes and the interface functions. The system has been widely applied to the multiple fields of the industrial control system, that has the advanced design ideas, and the characteristics of high real-time performance, convenient configuration, accurate computation.

Abstract： A general bottom layer driver design method is proposed, which is suitable for the development of chip drivers with different architectures. This is a top-down software structure written in C language. After the top-level framework is established, multiple people can collaborate to complete the development of each module driver, and then quickly enter the debugging stage, which greatly shortens the driver development process, saves time, and improves work efficiency. On the basis of this driver, the test program and application program of the chip can be further developed.

Abstract：Based on the hardware experiment platform with FPGA as the core, this paper designs and implements a complete software platform for the remote FPGA experiment system according to the requirements of remote configuration and control of FPGA, experimental results acquisition, equipment and user management in the remote experiment system, combined with image recognition, desktop client, Socket network communication，Web API server and other related technologies. The test results show that the software platform has good performance and accurate experimental results, which ensures a good experience for users when conducting remote computer hardware experiments, and also improves the flexibility of experimental teaching in schools.

Abstract： In order to meet the synchronous test requirements of quartz flexible accelerometers in parameter modeling system, a time-synchronized wireless test system is designed based on wireless communication technology. The system is designed based on the CC2530 wireless SOC chip. The hardware captures timestamps to reduce some delay factors. Least Squares Fit (LSF) is used to estimate and compensate for the relative clock drift between wireless nodes. The monitoring synchronization method is designed to reduce the number of times sync packets sent. The single-hop and multi-hop monitoring synchronization test results show that the average synchronization error of the four hops is 3.826 μs, the mean square error is 3.286 μs, and the average monitoring errors of the four hops are 8.193 μs and 5.241 μs, respectively. The system can meet the synchronous test requirements of the quartz flexible accelerometer parameter modeling application, and can be extended to similar applications.

Abstract：When CCS calls TI's own flash algorithm to operate on-chip flash, it takes a long time to erase, program and verify, which can easily lead to system hardware watchdog count overflow and system reset. The traditional processing methods require the user to modify the hardware or software each time the flash is programmed. There is a large human element to the finished system board. In order to avoid the potential risk of products being locked due to the above-mentioned human factors, this paper provides a set of general-purpose, easy-to-implement flash algorithms that can exclude human factors. Instead of the flash algorithm that comes with CCS, it adds a pointer to the dog feeding function or other user-customized function pointers, so as to solve the embarrassing situation that the user cannot feed the dog or other functions when CCS operates flash. The test results show that the flash algorithm can add custom functions and dog feeding functions according to user needs, avoid changes to software or hardware, and improve the stability and reliability of the system in the debugging stage.

Abstract：This paper analyzes the current situation and limitations of radio frequency identification reader interface protocol test technology at home and abroad, and discusses the necessity of further developing the reader interface protocol test system. It discusses a reader interface protocol test system, including test device and software design and test method design. Through the development of reader interface protocol test device and software, it fills the gap in the field of military radio frequency identification reader interface protocol compliance testing. Through the development of reader interface protocol test method, it further improves the military RFID standard system, provides technical support for the application of military RFID equipment, and improves efficiency, stability and safety.

Abstract： Nowadays, the competition in the field of domestic semiconductor largely depends on the competition of semiconductor equipment and core components, such as RF power supply, impedance match, etc. The anti-interference ability and stability of the equipment are important criteria for judging the quality of equipment. For the impedance match equipment, the accuracy, stability and fast response of the control loop are important indicators that determine the performance of this product. Therefore, this paper designs a hardware circuit about amplitude-phase signal detection loop based on the related principles of capacitor voltage sampling, current transformer sampling and op amp circuit, and uses the ADS2021 to obtain the analog quantity Urms representing the impedance matching degree between RF power supply and load. The simulation experimental results show that a larger Urms value indicates a greater mismatch between the RF power supply and load, and a smaller Urms value indicates a better match between the RF power supply and load. At the same time, by changing the input power of the bus signal, it is obtained that the Urms has a high linear correlation with input power, so the detection loop has good stability and can be applied to the detection of impedance match.

Abstract： In complex and harsh electromagnetic environment, radar will be interfered, resulting in unrecognizable target or target loss. Frequency hopping is an effective measure to deal with electronic interference from enemy or natural environment. Frequency hopping methods include blind hopping, pseudo-random frequency hopping and adaptive frequency hopping. Aiming at adaptive frequency hopping technology, a three-dimensional random matrix adaptive frequency hopping algorithm is proposed. Firstly, the time set matrix, the frequency point interference amplitude set matrix, and the available frequency point set matrix are established. On the basis of these, the three-dimensional random matrix mathematical model is generated, and then the implementation method is given. Combined with practical engineering applications, the effectiveness of the algorithm is verified. The results show that the algorithm improves the pulse compression ratio of radar to a certain extent, greatly improves the anti-jamming ability of radar in electromagnetic environment, and is of great significance to improve the radar target recognition ability.

Abstract：The current/frequency conversion circuit is mainly used to measure the output current of accelerometers in inertial navigation system, and at present the typical current frequency conversion circuit consumes so much power, which limits its application scope. Shunt circuit is usually used to design low power current/frequency circuit, however this method will bring large zero bias error, which affects the accuracy of circuit measurement. The reasons that affect the zero bias of low power current frequency conversion circuit are analyzed, and a method to realize zero offset compensation by single chip microcomputer is presented, it only needs to load the program several times to realize iterative compensation, through coarse compensation for three typical temperature points and fine compensation for the whole temperature range, and it is convenient to realize high-precision zero offset compensation. The experiments show the zero offset of the designed product at whole temperature is less than 0.15 Hz, and the power consumption is 1/5 of that of typical circuit, which has high engineering application value and good market prospect.

Abstract：In order to solve the problem that SC weather radar needs to generate products after volume scanning, a real-time echo display scheme for SC weather radar is designed and tested at Sanya radar station. In the three schemes, the scheme that FTP simultaneously distributes elevation by elevation files and standard format base data is selected, and the original RDA software and FTP transmission software are redeveloped to enable it to generate and transmit elevation by elevation files and standard format base data at the same time. The generation time of products No. 19, 20, 37, 38, 61 and 110 generated by SA standard base data, SC elevation file and SC standard base data are compared. The results show that: (1) The generation time of products using SC elevation file is equivalent to that of SA radar; (2) The generation time of typical PPI products using SC elevation file is greatly reduced compared with similar products using SC base data; (3) The time of generating derivative products by using SC elevation file is equivalent to that of generating similar products by using SC base data. This design scheme is universal and can be extended to other types of weather radars in China to shorten the generation time of some products and improve the forecasting time of short-time heavy rainfall and storm.