Abstract：The demand for power sensors is growing rapidly, and the demand for energy supply is becoming more and more obvious. The disadvantages of the traditional single-battery power supply scheme are becoming more and more obvious. At the same time, with the development of kinetic energy, temperature difference, photovoltaic, electromagnetic and other energy collection technologies and integrated circuit technologies, new ideas are put forward for the energy supply of power sensors. Promoting energy harvesting technology and power sensor coupling in smart grids can improve social benefits and form a win-win situation for all parties. This paper firstly analyzes the urgent demand of smart grid for energy harvesting technology. Secondly, the development status of energy harvesting technology and related integrated circuits is analyzed. Finally, according to the shortcomings of energy harvesting technology in smart grid, the research direction of energy harvesting technology suitable for smart grid is proposed.

Abstract：To improve safety during ordinary railway catenary maintenance operations and the anti-interference of transmission information, a set of ordinary railway catenary condition monitoring system based on artificial intelligence robot technology and LoRa communication technology is designed. The system is mainly composed of a robot that can inspect electricity, hook ground wire and return line in the job site as a data acquisition device and the pattern constitution of transmitting and processing status information outside the job site and realizing visualized monitoring at the terminal. After the monitoring and simulation test, the anti-interference results of the robot data collection are in line with expectations. Meanwhile, the data transmission between the LoRa gateway and the cloud server is normal, and visualized monitoring of the status information from the PC side is realized.

Abstract：Aiming at the problem that the accuracy of text emotion classification of online comment is not high, an improved model based on BERT and bidirectional gated recurrent unit (BiGRU) is proposed. The word vector representation is carried out by using the BERT model which can represent the rich semantic features of the text. The classification effect of the model is improved by combining the BiGRU neural network which can retain the text context related information for a long time. On this basis, the attention mechanism is introduced, to highlight the weight of emotional words which can better express the classification results in the text, and improve the accuracy of emotional classification. The above model was tested on Acllmdb_v1 data set and hotel reviews data set, which are public data set. The experimental results show that the model achieves good performance in both Chinese and English text emotion classification tasks.

Abstract：The open core protocol (OCP) bus can be applied to decouple IP core functions and the interfaces to realize the plug-and-play function. Aiming at the synchronization problem when the OCP is connected to asynchronous clock domain, a lightweight synchronization interface is developed, which not only synchronizes the control signals but also reduces the hardware consumption caused by data buffer across the clock domain. In view of the scalability of the point-to-point OCP bus, the enhanced clock-domain-crossing OCP bus is deployed on the AMBA High-performance Bus (AHB)，which is interconnected by the shared bus, in order to realize multi-master-slave multi-clock transmission. It is proved by simulation that the enhanced clock-domain-crossing OCP-AHB bus can transmit data correctly, which is able to be rapidly deployed in the next step.

Abstract：With the rapid development of IoT technology and the widespread deployment of IoT devices, the issue of information security has become increasingly prominent. Cryptography is the key core technology to ensure information security, but the traditional cryptographic algorithm adaptation scheme is difficult to balance performance and flexibility, this paper proposes a cryptographic instruction extension scheme to achieve a good balance between the two scheme. Firstly, we analyze the computational aspects of the AES algorithm, and propose a cryptographic instruction extension and acceleration engine design scheme by combining the Hummingbird E203 processor architecture; then we complete hardware and software implementation, build an RTL-level simulation environment and an FPGA board-level verification environment; finally, we perform experimental verification and comparative analysis. The experimental results show that the proposed scheme can achieve about 700% acceleration ratio with only nearly 2% increase in hardware resources, which has high energy efficiency and can be applied in resource-constrained situations such as IoT.

Abstract：As the power supply of IC chips tends to be low-voltage and high-current, the design of the Power Delivery Network(PDN) of the micro-system module is becoming more and more important, based on 2.5D Through-Silicon-Via(TSV), inverted solder, High Temperature Co-fired Ceramics(HTCC),3D stacking, etc. Chip currents generate output noise through PDN interconnects, which must provide a better signal return path with low-impedance to keep the supply voltage between chip pads constant within a small tolerance range, usually 5% or less. Based on Chip Package System(CPS), the paper proposes a design and optimization method for the micro-system module PDN based on the three-level synergy of TSV silicon substrate, HTCC cases and PCB. This paper expounds the DC design and AC impedance design respectively, and using the Chip Power Model(CPM) combined with the time domain analysis, realizes the low impedance design of the power ripple on PDN.

Abstract：This paper presents a high accuracy and low phase variation 5-bit digital controlled attenuator (DCA) for Ka-band phased-array system applications. The DCA adopts three attenuation topologies of T-type, simplified T-type and Π-type topologies with embedded switches to design the basic attenuation cells, and realizes the attenuation dynamic range of 15.5 dB and the minimum attenuation step of 0.5 dB. Amplitude calibration technology is used in signal path, which can effectively reduce the increase of attenuation amplitude error caused by process fluctuation and enhance the robustness of circuit design. At the same time, capacitance compensation technology is used in T-type and Π-type attenuation topologies to improve their high-frequency attenuation performance and achieve low phase variation. Based on 65 nm CMOS process, the proposed DCA is optimization simulated, taped out, measured and verified. The core area of the chip is 500 µm×150 µm. The measurement results show that within frequency range of 25~35 GHz, the insertion loss of reference state is 6.54~8.6 dB, the input/output return loss corresponding to 32 attenuation states is better than -15 dB, and the amplitude error RMS and phase error RMS are 0.12～0.26 dB and 1.02～2.07o, respectively.

Abstract：As a mainstream bus protocol, the application scenarios of PCIE(Peripheral Component Interconnect Express) bus are more and more abundant, and the peripheral devices connected on the bus are also increasing. On both desktop and embedded general-purpose processors, the number of PCIE controllers is limited, and the functions of many PCIE bridge chips are also very limited. A polymorphic PCIE expansion bridge chip is designed to extend the processor's processing capability. The chip's multi-channel, high-throughput, polymorphic attributes effectively make up for the shortcomings of the traditional PCIE processor. The scheme determines the feasibility and the stability of the chip through the post-silicon verification method, and also provides a designed idea suitable for the follow-up PCIE extended chip.

Abstract：Aiming at the requirement of constant force control precision and response speed of grinding robot in grinding process, a PID-type fuzzy controller for constant force actuator optimized with particle swarm optimization is proposed. A new PID-type fuzzy controller is designed to reduce the number of design rule bases. The variable weight comprehensive fitness function is proposed, which combines the absolute value of the error integral and the absolute value of the control signal integral to optimized the comprehensive performance of the PID-type fuzzy controller, while reducing overshoot quantity and steady-state error. The adaptive inertia weight strategy is used to speed up the iteration speed of the particle swarm, the scale factors of the PID-type fuzzy controller is optimized with the particle swarm algorithm. The simulation results show that comparing to the traditional PID the PID-type fuzzy controller with particle swarm optimization has a smoother output curve, its response speed is increased by 10%, its adjustment time is shortened by 14%, and it has no overshoot and no oscillation, which improves the control precision of grinding force.

Abstract：At present, in chip design companies, sample verification is carried out for mass production chips. Most of the traditional sample verification methods design corresponding test equipment based on the characteristics of the chip, and then test the chip samples one by one through the test fixture. Different chips will design different test equipment. In this paper, an adaptive and synchronous test sample verification platform scheme is proposed, which can not only test multiple chips at the same time, but also test chips with different interfaces. It can not only test the reliability experiment, but also test other functions, which greatly saves the maintenance cost of test equipment and improves the test efficiency.

Abstract：The seeker position calibration system is a frame type angle measuring device, and the static balance of its inner ring components directly determines the stability of the product performance. This paper makes full use of the advantages of the pressure sensor, carries out analog-to-digital conversion and data acquisition through HX711, displays the collected results on the LCD screen, and analyzes and processes the data. A device that automatically measures the static balance of the frame in the seeker position calibration system is designed. The device uses four embedded pressure sensors to realize that if the product is unbalanced when placing the inner frame of the product on the balancing tool rest, one end of the inner frame will press on the balancing device due to gravity. By displaying the size of the pressure value and fitting with the least square method, the relationship between the pressure value and the thickness of the increased or decreased shims of the product is realized, and the universality of the relationship is verified. One calculation can realize the accurate adjustment of the static balance of the frame in the seeker position calibration system.

Abstract：In order to solve the wideband IQ imbalance problem, a compensation design based on blind equalization is presented. An improved equalizer design is used in high speed demodulator, which computes IQ imbalance compensation filter by adaptive blind equalization technology, therefore the equalizer has the ability to compensate IQ imbalance and reduce inter symbol interference. In engineering application, this method does not increase additional resource consumption and hardward design complexity. The test result shows this method can compensate IQ imbalance effectively in multiple modulation modes.

Abstract：PBFT algorithm has some problems such as high communication complexity and performance decrease with the increase of nodes.This paper introduces multicast communication applied to RPBFT(role-based Byzantine consensus mechanism),this paper proposes a WRPBFT consensus algorithm with low communication complexity, which can allocate consensus nodes dynamically and does not degrade with the increase of the number of nodes.WRPBF divides nodes into consensus nodes and candidate nodes, and dynamically selects consensus nodes participating in consensus for multicast communication.Experiments and analysis show that WRPBFT consensus algorithm can dynamically achieve node partitioning and selection, and WRPBFT algorithm has higher throughput, lower delay and lower communication complexity compared with traditional PBFT algorithm.

Abstract：Aiming at the problems of low real-time performance, large redundancy, and inability to recover from faults in the communication between heterogeneous processors in the Space Telecommunication Radio System (STRS),the distributed data distribution service (DDS) middleware technology is introduced into the STRS architecture to realize the publish/subscribe mode-based communication middleware between the STRS heterogeneous processor waveform application components.Under the premise of being fully compatible with the STRS standard specification, it effectively improves the real-time performance of the communication system based on STRS, reduces the complexity and redundancy of the system, improves the development efficiency, saves the development and maintenance costs of the system, and realizes the dynamic refactoring of global and local modules.

Abstract：Time-domain randomization makes side channel attacks difficult to align side information by inserting random delays, redundant operations, and variable frequency clocks, thereby reducing the efficiency. Aiming at the problem of time-domain misalignment of energy traces caused by time-dimensional randomization technology, this paper proposes an anti-time-domain randomization method based on Trend Series Dynamic Matching (TSDM). This method analyzes the distribution of attack interest points, and describes the shape feature of the energy trace, compresses the dimension of the energy trace, preserves the sensitive information of energy attacks, suppresses the negative impact of noise, realizes the dynamic alignment of data, and improves the efficiency of side channel attacks. This paper, conducts method verification and performance analysis for three typical time randomization techniques. Compared with various classical alignment methods, this method has better data alignment effect in different noise environments, and the number of energy traces required for the success rate of energy attack to reach 100% respectively decreased by 23.8%, 24.2%, 11.3%.

Abstract：With the rapid development of civil aviation industry, more and more tasks are required to fuel the aircraft. At present, the aircraft refueling task is mainly assigned to the refueling personnel by the dispatcher of the fuel supplier, one by one, according to the flight refueling information, after receiving the task, the refueling operator drives the refueling vehicle to the aircraft for refueling operation. The realization of electronic, information and big data analysis technology has gradually opened up the information channel of refueling system. Based on a large number of multi-dimensional data, according to the characteristics of aircraft refueling operation, an intelligent dispatching method for aircraft refueling vehicle is provided, which can automatically assign refueling tasks, avoid flight delays and the resulting serious waste of resources and can reduce the workload of the dispatcher. Through the intelligent dispatch, refueling resources can be reasonably allocated to ensure the timely refueling of aircraft and smooth completion of flight support tasks.

Abstract：Network security has become a major issue related to national security and development and the vital interests of the people. The network security situation is also becoming increasingly severe. Network attacks endanger political security, social stability, economic development and cultural construction. International competition in cyberspace is in the ascendant. This system identifies and warns potential network risks from multiple dimensions, takes effective measures to block the attack before the attack is successful, which can ensure the stable operation of the information system.

Abstract：This paper discussed a CQFN shell in 7 mm×7 mm×1.2 mm which was based on HTCC technology for the miniaturization needs of millimeter wave microwave device packaging. In order to solve the problem of electromagnetic leakage during high frequency signal transmission of ceramic shell, the shielding ground hole is designed and optimized from different directions of signal transmission by using HFSS. Through simulation comparison, the shielding relationship between ground hole and electromagnetic signal in different areas is discussed and summarized. The results show that the RF pin could be used in the band from 0.1 GHz～40 GHz. The insertion loss of RF ports is less than 0.65 dB. The VSWR of the package is less than 1.50.

Abstract：A compact and low-cost transceiver front-end module is designed for point-to-point radio communication system in this paper. The transceiver module contains one receive channel, one transmit channel, two groups of LO synthesizers in limited area. The module is implemented using devices with Surface Mount Technology (SMT) and multi-layer PCB for excellent performance, low cost and easy assembly process. This paper firstly introduces the architectures of transceiver front-end, then gives detailed design of key passive and active circuits, and analysis of main performance index. The assembled module is measured, it can transmit more than 20 dBm and receive -60 dBm to -20 dBm RF signal. The output IF signal of receive channel is -10 dBm constantly.

Abstract：This paper proposes a SoC chip clock scheme and designs two layout schemes. By analyzing the causes of electromagnetic interference and judging the simulation results of the isolation degree between signals, the risk of inter-signal interference in the layout scheme can be identified in advance, which will deteriorate the output analog signal quality of SoC chip. In this paper, the simulation results of the isolation degree between signals are used to guide the layout design, and the optimization layout method is given to improve the performance of the isolation degree between signals. The test results show that the simulation results of the two layout design schemes are consistent with the actual results, which verifies the accuracy and reference of the simulation results.

Abstract：To overcome conventional VHF power amplifiers′ shortcomings of bulky size, low efficiency, and EM interferences due to output harmonics, a miniaturized, high power and high efficiency power amplifier working in VHF band is designed by using lumped parameter elements in this paper. In this work, multiple series LC resonators are introduced into the output matching circuit of the power amplifier, with effectively suppressed harmonic components, improved linearity and efficiency, as well as improved electromagnetic compatibility of the power amplifier. Finally, the physical test results show that the saturated output power of the power amplifier is about 44.9 dBm, the saturated power gain is 14 dB, and the power added efficiency is 62%. With its excellent performance, it has good practical engineering application value and can be widely used in aviation, aerospace and other fields.

Abstract：Due to highly customized feature, once the program of FPGA system is fixed, it is extremely difficult to keep maintenance. The system uses MicroBlaze processor to implement TCP/IP protocol stack, realizes the real-time online update of modules in FPGA system by upgrading files through Internet, and completes remote command injection and data upload for remote debugging. Based on ethernet mode, the system can be separated from the traditional debugging method, realize remote equipment debugging without manual participation on the equipment site, and reduce maintenance costs. The system has the characteristics of reliability, fast configuration, and no need to restart using this method, and can also be used in other directions such as cloud computing and real-time simulation.

Abstract：Non-intrusive load monitoring (NILM) technology is only based on the current and voltage information of the main entrance of home power supply to obtain the electrical information of indoor electrical equipment. Improving the accuracy of load identification is of great significance to optimize the energy structure, improve the efficiency of power utilization and reduce energy consumption. Firstly, the normalized current signal is decomposed by using variational mode decomposition (VMD), and then the correlation coefficients between each component and the original current signal are calculated. The two components with the largest correlation coefficients are selected as the load characteristics and input into the trained LSTM neural network for identification. The test results of an example show that the recognition rate of this method is up to 99% on public data set PLAID and 96.6% on laboratory data set, which proves the effectiveness of this method.

Abstract：High efficiency, high switching frequency and digital control are the research hotspots in the field of switching power supply in recent years. In this paper, the design strategy of synchronous buck digital power supply based on average current mode is discussed. Combined with theoretical analysis and Matlab/Simulink software simulation, the loop control design conditions of synchronous buck are analyzed, two design methods of voltage and current double closed-loop PI compensator are described in detail, and a voltage loop control design method based on the simplification of the current loop is proposed. The high-performance mainstream MCU chip STM32G474 is used as the control core, and the high-resolution PWM technology is used to realize a switching frequency of 200 kHz, and the architecture of the system software is constructed with the idea of interrupt event triggering and state machine operation. The simulation and experimental results show that the system has the characteristics of high efficiency, fast dynamic response and strong robustness, which provides a reference for the general design of digital power supply.

Abstract：In order to solve the problems of large quantity of mass satellite production, short cycle, and shortage of personnel, a new type of satellite automatic test system is proposed. The hardware architecture of the system is composed of a comprehensive test server, remote-control telemetry ground inspection equipment, and an automated test terminal. The software architecture is based on the joint development by TestStand and LabVIEW. The code module edited by LabVIEW realizes the function of creating information channels with remote control and telemetering ground inspection equipment, server databases, etc. Following the operation process of the test instructions， TestStand edits the automated test sequence, realizing the highly automated processes such as remote control instruction sending, telemetry data interpretation, process monitoring, test report generation, and multi-satellite parallel testing. The test sequence of this architecture has high flexibility and strong versatility, and can meet the demands of a variety of satellite automated test conditions. It has been successfully applied in the mass production and testing process of Jilin-1 Gaofen 03 series satellites. While greatly improving the test efficiency, it saves a lot of manpower.

Abstract：As the core basic equipment of the industrial control system, PLC is widely used in important national infrastructure and various industrial control fields. It is the top priority to realize the transformation from made in China to intelligent manufacturing in China to control the industrial site independently, safely, stably and reliably. In this context, domestic PLC has a strong and rapid development momentum, especially in high-end application fields such as aerospace, national defense and military industry. There is an urgent demand for domestic high-performance PLC products. Due to the late development and less application of domestic large-scale PLC, how to evaluate the performance of domestic PLC and whether it can meet the industrial field application is lack of unified standards and evaluation methods. In view of this problem, a set of comprehensive test platform for domestic high-performance PLC is designed. Two mainstream domestic PLCs are selected for quantitative comparison of horizontal parameters with a foreign mainstream product. The key performance indicators and evaluation methods of high-end PLC are proposed. By building a hardware in the loop virtual industrial scene, the ability of PLC to meet the application is tested and compared. The effectiveness of the test platform is verified through a test example, which provides a basis for the evaluation of domestic high-performance PLC Choose to provide reference, promote the normalization, standardization and scale of domestic PLC, and help the high-quality development of Industry 4.0.