Abstract：In order to effectively and automatically select communication links and reduce the global risk of communication, this paper proposes an automatic selection method of power 5G communication links based on risk balance. First of all, taking the lowest global risk and the highest service channel availability as the objective functions, and taking the channel bandwidth, delay and packet loss rate as the constraints, the automatic selection model of power 5G communication link is established. The business risk degree is evaluated by taking the business risk degree and the network business risk balance degree as indicators. According to the degree of risk, the communication services are sorted from the lowest to the highest. The automatic selection model of communication links is used to give priority to the selection of links for the services with lower risk. The automatic selection model of communication links is gotten through quantum genetic algorithm, to obtain the lowest global risk and the highest service channel availability value, and complete the automatic selection of communication links. Experiments show that this method can effectively obtain the risk degree of each communication service, automatically select the communication link, reduce the global risk degree, improve the availability and bandwidth of the service channel, reduce the channel delay and packet loss rate, and achieve better risk balancing effect.

Abstract：Based on a large amount of energy consumption, performance indicators and base station configuration data that can be collected in the 5G network management system, this paper uses machine learning algorithms to establish an energy consumption calculation model for common AAU equipment on the existing network, and verifies the accuracy of the model. The test results show that the accuracy of the energy consumption calculation models established by this scheme for several common AAU devices in the existing network has reached more than 97%, which fully proves that the energy consumption modeling method has high practicability and promotion value.

Abstract：The combination of high-precision indoor positioning technology and small cell communication network can effectively reduce the cost of deployment and operation and maintenance, and provide important solutions for various positioning application scenarios. Firstly, the common positioning technologies were summarized and compared, and the feasibility of different positioning technologies in small cells was further analyzed. Then, an indoor positioning scheme based on small cells was proposed, and its architecture was described in detail. The scheme had the advantages of easy expansion of transmission network, flexible transmission path and open positioning interface. Finally, the application scenarios and positioning processes of the two positioning technologies selected based on this scheme were presented. The results show that the indoor positioning scheme based on small cells can meet the positioning needs of the consumers and a wider range of vertical industries, which has great significance and application prospects.

Abstract： In order to ensure the shielding performance of the shielding cavity of the 5G equipment in the converter station, and to clarify the rule of the influence of the number and size of the cavity aperture on its shielding effectiveness, this study constructs the electromagnetic shielding effectiveness analysis model of the shielding cavity under the HFSS simulation environment, and analyzes the spatial distribution characteristics of the electromagnetic radiation of the 5G microstrip patch antenna. Then, the influence of the slot height of the cavity on the shielding effectiveness is analyzed. It is found that the shielding effectiveness under different slot diameters increases with the increase of the slot height, and the electromagnetic radiation intensity decreases from 24.81 dB at 0.5 mm height to 16.29 dB at 2.0 mm height. At the same time, the analysis shows that the shielding effectiveness decreases with the increase of the slot diameter, and the maximum radiation intensity increases from -15.13 dB at 1.0 mm diameter to 49.87 dB at 7.0 mm diameter. Finally, with a certain proportion of the aperture area, it is found that the shielding effectiveness of the shielding cavity is continuously improved with the increase of the aperture number and the decrease of the aperture diameter.

Abstract： This paper briefly reviewed the history of China-made PLC and presented the domestic market situation of PLC, mainstream products of China-made PLC, PLC standards, and relevant national policies. And the existing gaps between foreign and domestic PLC products were analyzed from some aspects including PLC performance indicators, logic programming software, network protocols, product categories, and standardization. Then this paper tried to propose a solution to develop the China-made PLC, combining the characteristics of PLCs of European, American, and Japanese systems. Led by the key projects in the fields of defense industry and critical infrastructure, the China-made PLC will be developed and used based on the domestic independent software and hardware.

Abstract：In order to achieve the goal of reducing taxi operating losses, the quantification of losses should first be completed. The average time cost of waiting for drivers is quantified by queuing theory, and the "Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS)" model improved by entropy method quantifies the possibility of carrying passengers in different periods. The possibility is used to complete the quantification of potential loss parameters of no-load. Through the quantitative indicators, a taxi operation loss model is constructed. Based on this model, an operational decision-making scheme to reduce losses is proposed. Finally, the verification of flight and taxi data quoted from Luoyang Beijiao airport proves that the decision-making method of this paper can effectively reduce the operational loss of taxis.

Abstract： In order to solve the problem of perception and no decision-making in the agricultural modernization of Xinjiang Corps, an image classification method (TL-DA-SE-CNN) based on attention mechanism module (SENet) and convolutional neural network hybrid model transfer learning is proposed. This method selects four different CNN models for weight acquisition, including VGGNet, ResNet, InceptionNet and MobileNet. The model uses the SENet classifier instead of the fully connected layer of the convolutional neural network, extracts the structural high-order statistical features of the image for topic classification, and uses the BP algorithm to adjust the parameters, with a classification accuracy of 98.20%. Experimental results show that the technology of combining CNN with transfer learning, data augmentation and SENet improves the performance of livestock image classification, which is an effective application of convolutional neural network in farm automation clustering.

Abstract：A debugging system based on JTAG interface is proposed in this paper to provide an effective and convenient debugging method for domestic processor chip. The debugging system, which is based on JTAG standard, simplifies the design of the on-chip debugging hardware module. With little hardware overhead, simple and high performance design of the debugging instruction, it is used to realize the debugging interrupt, breakpoint and watch point setting，single step running，register or memory read and write, and other base debugging functions, as well as sence protection and recovery, trace buffer, instruction insert execution and other advanced debugging functions. After passed the actual chip testing, the debugging system has compatibility with JTAG protocol, comprehensive functions, high performance, simple structure, convenient for operation and other features.

Abstract： Based on V93000 ATE, a test system using external power boost conversion module and programmable gain instrumentation operational amplifier was designed to meet the requirements of mass production of multi-sites ADC with large-swing and high-precision. The theoretical and test results show that the ADC test system can generate 29.4 V ramp and 29.1 V sine wave signals with peak to peak value. The SNR of test signal is better than 105 dB and THD is better than -103 dB, which can meet the test requirements of 16 bit high-precision and ±10 V high input swing multi-sites ADC.

Abstract： Based on laterally diffused metal-oxide-semiconductor (LDMOS) devices, a broadband RF power amplifier for L-band was developed. The amplifier is composed of two stages of amplification and cascade connection. In order to achieve broadband and good output standing wave, the final stage power amplifier adopts a balanced topology circuit structure; the inter-stage matching network uses the microstrip line and capacitor hybrid matching method to achieve broadband matching. The final measured data are as follows: the frequency covers 0.96 GHz~1.4 GHz, the overall output power of the power amplifier reaches 50 dBm (100 W), the power gain is greater than 30 dB, and the efficiency is greater than 45%. The power backoff is 8 dB, and when the output power is 42 dBm, the adjacent channel power ratio (ACPR) is -40 dBC. The indicators show that the power amplifier module can be well used in radar and wireless communication transmitters.

Abstract：Aiming at the problem that the noise information is fixed when the extended Kalman filter(EKF) algorithm estimates the state of charge(SOC) of lithium-ion battery, resulting in low estimation accuracy, an adaptive extended Kalman filter(AEKF) algorithm with automatic matching of noise information covariance is proposed. Firstly, the parameters are identified based on the dual polarization (DP)equivalent circuit model of the battery, and an accurate equivalent model is established. Then, the variation of noise covariance matrix of EKF filtering algorithm and AEKF filtering algorithm and the estimation effect of battery SOC are compared under dynamic stress test(DST) conditions. The results show that AEKF filtering algorithm has higher estimation accuracy. Finally, several groups of different SOC initial deviations are set to verify the strong robustness of AEKF filtering algorithm in estimating battery SOC.

Abstract：Thermostat plays an important role in people′s production, life and scientific research, such as the measurement and compensation of temperature compensated crystal oscillator parameters. This paper designs and implements a software-controlled portable thermostat control system with low fluctuation. This thermostat uses a method to control the power of ceramic heating chip or semiconductor refrigeration chip by dynamically adjusting the working voltage, so as to realize the constant control of temperature. This thermostat has advantages of low temperature fluctuation, supporting temperature curve software control, small and portable. After testing, the temperature control range of the thermostat realized in this design is from -15 ℃ to 80 ℃, and the temperature fluctuation in stable state is ±0.2 ℃. In addition, compared with the thermostat using pulse width modulation control method, this thermostat has lower system electromagnetic radiation and higher refrigeration efficiency. Under the same power input condition, the temperature can be further reduced by 12.9 ℃.

Abstract：Aiming at the problem of high sidelobe in the pattern of general uniform planar array, the traditional genetic algorithm is used to arrange the uniform array in two-dimensional sparse distribution,which effectively reduces the sidelobe level, but the convergence speed of genetic algorithm is slow and easy to fall into local optimal solution. Because chaotic optimization algorithm has the characteristics of randomness, ergodicity and regularity, introducing chaotic optimization algorithm into genetic algorithm and initializing population by chaotic sequence can improve the convergence speed of genetic algorithm and the ability to obtain global optimal solution. Therefore, a genetic algorithm based on chaotic optimization algorithm is proposed and applied to the design of two-dimensional planar array antenna.The algorithm optimizes the layout of antenna array. The simulation results show that the convergence speed of chaotic genetic algorithm is improved and the sidelobe level of array antenna is further reduced,which shows that this method is feasible.

Abstract：In order to solve the problem of rapid iterative interaction between multiple versions and services under the cloud native architecture, a full-link grayscale publishing technology scheme for cloud native is designed. Using traffic dyeing technology, an isolated environment from the gateway to all back-end business services is constructed to achieve grayscale publishing supporting multiple versions and multiple services. Flexible and rich gray scale strategy is designed to realize precise control of gray scale publishing. Full-link grayscale publishing capability is implemented through gateways and sidecar module that are relied upon by all back-end business services. Engineering practice shows that this scheme can effectively isolate the environment logically, provide precise control and full link capability, and effectively reduce the impact of demand release and improve the efficiency of online release.

Abstract：Aiming at the security of some key data storage in network security system, this paper designs a data security storage and processing system based on blockchain technology. Firstly, taking the thermal power plant as an example, the architecture of its control system and network security system is analyzed. Then, based on the architectural characteristics of thermal power plant, a specific secure data storage and processing system is designed, the architecture of application deployment is given, and the data secure storage and processing process is analyzed. Finally, the designed system is tested and verified. The verification results show that the system is correct and effective for data anomaly detection, and has strong practical application and popularization value.

Abstract：For the effective, convenient and unified management of materials about military logistics, this paper studies and designs a monitoring system for monitoring the presence of items. The system uses the motherboard of Raspberry Pi to collect the level signal of the infrared sensor about the presence of the items and the images taken by camera of the items. Then it converts the data into the general data frame, and sends the frame to the data processing module through the source-specific multicast. Finally, the image recognition based on the convolutional neural network is used to judge the correctness of the item, and display its status in real time through the monitoring interface. It has been verified that the system can ensure the real time of data acquisition and the accuracy of identifying items. It has strong practicability.

Abstract：At present, the power output of electric vehicles is mainly derived from power batteries, whose State of Charge (SOC) represents the remaining power of batteries. Accurate estimation of SOC is of great significance for the safety of battery use . Butterfly Optimization Algorithm (BOA) was improved and used to optimize BP neural network to estimate SOC of power battery, which solved the problems of long training time, slow convergence, low accuracy and easy to fall into local optimal solution. At the same time, the global search speed is improved, voltage and current are selected as input variables, SOC as output variables, and the weight and threshold of neural network are adjusted according to the size of error. Simulation results show that the error rate of SOC estimation results obtained after optimization is controlled within 1.1%, and this method has better robustness and faster optimization speed.

Abstract： In order to realize real-time and multi-dimensional control of substation risk early warning, a substation risk early warning and control system based on edge computing is designed. The sensing layer collects the operation data of substation equipment through the Internet of Things equipment, and transmits the data to the edge layer through the communication between edge equipment and external equipment. After receiving data, edge layer parses the data through the communication protocol to complete data consistency, and packages cloud services downstream data sending to the cloud computing layer. The layer uses cloud computing to compute and analyze substation equipment data and transmit it to the application layer. The application layer adopts support vector machine (SVM) algorithm to establish a risk early warning mechanism, divide the risk data attribute, and obtain warning value. Based on this, the risk situation corresponding to the early warning value is predicted by the logistic regression risk prediction model, and the probability of failure is obtained to complete the substation risk warning. Meanwhile, the early warning results are presented on the visual interface, and the corresponding risk warning is controlled and controlled by the control channel. The test results show that the communication jitter results of the system are all below 15 ms, which can realize the calculation of risk probability, obtain the number of various risk warnings of the substation, have good work energy efficiency, and realize the multi-dimensional risk warning and control of the substation.

Abstract： Portable recording equipment for law enforcement plays a key role in the law enforcement process of evidence, but because it is worn on the body, the recorded video shakes seriously, which affects the subsequent viewing and processing. In order to improve the anti-shake effect of portable recording equipment for law enforcement, a scheme combining two-axis mechanical anti-shake and video anti-shake was proposed. At the same time, a dynamic calibration scheme was proposed to improve the stability of anti-shaking effect of law enforcement instrument in complex environment. In order to solve the IMU data distortion problem caused by the restriction of collision to the two-axis mechanical anti-shake of the instrument, a fast recovery scheme was proposed.

Abstract：The active and high-frequency harmonic current are compensated by introducing power electronics converter of the flexible grounding, which achieves full compensation of the single-phase to earth current to achieve a fast arc extinction. Thus, the flexible grounding is promising in the distribution network. Regarding the active and harmonic current control, this paper proposes to design the controller in the discrete-time domain based on the discrete model of the plant, which deals with the one-sample delay directly. The discrete complex current controller is developed which achieves AC current regulation in the stationary reference frame without steady-state error. With regard to the proposed controller, a tuning method based on critical damping is proposed based on the discrete root locus analysis, which avoids overshoot and optimizes the settling time. The simulation results have verified the effectiveness of the proposed controller and the tuning method.

Abstract： In the domain of data storage applications such as radar, communication, and electronic countermeasure, in order to meet the requirements of writing large capacity SATA Solid State Disk Array (SSD) continuously, steadily and high-speedily, a RAID0 storage array system based on FPGA is designed. This system takes FPGA as the data processing core, which realizes functions such as data receiving, data buffering, data transmission, and data storage. According to the garbage collection characteristic of SATA SSD, logic function modules are designed and optimized. The repeatable tests show that sixteen SATA SSD has the storage capacity up to 32 TB, writing bandwidth up to 5.5 GB/s, and the ability to work steadily for a long time in the industrial environment.

Abstract： Common-used Convolutional Neural Networks (CNNs) contain billions of multiplications, which is the bottleneck of hardware implementation of CNNs. To reduce energy cost of multiplier, an energy-efficient radix-4 Booth encoder multiplier is proposed. By improving the partial product module, the compensation bits in conventional multipliers are eliminated, which reduces the delay and energy cost of multiplier. Post simulation indicates that the proposed multiplier reduces the area, delay and energy cost by 5.2%, 6.3% and 10.8% respectively. The proposed multiplier can be used in neural network accelerators and breaks the energy efficiency bottleneck.

Abstract： Chronic diseases which include cancer diseases, respiratory diseases and cardiovascular & cerebrovascular disease become the main diseases affecting the health of Chinese residents, and affect the improvement of healthy life expectancy of Chinese residents. Therefore, using the portable and real-time medical equipment to track and detect patient′s daily physical conditions is very important. This paper has a comparative study and designs two typical ECG devices, one is the sport monitor-SenseON which is a small and stylish, the other one is the medical monitor-CareON. This paper also discusses the future trends.

Abstract：The paper proposes a question and answer robot for government knowledge based on domestic PKS environment, which uses deep learning model to work in pipeline mode for multi-source heterogeneous data. The system consists of five parts: schema design, knowledge extraction, knowledge fusion, knowledge storage and knowledge update, automatically completes the establishment of knowledge base, automatically classifies user questions, and recognizes intention. Knowledge map retrieval, analysis and computing technology are used to form answers for users. The robot system can greatly improve the work efficiency of government staff.

Abstract：The self-control cloud collaborative office platform based on network, cloud and end architecture adopts the "network, cloud and end" architecture mode, and comprehensively uses new technologies such as cloud computing, microservice, SDN and cloud security to create a networked and intelligent collaborative office cloud service system with advanced technology, comprehensive functions, convenient use, flexible deployment, intelligent operation and maintenance, compatibility and adaptation, security and reliability. Through the construction of comprehensive network system, it can achieve security, credibility, Flexible and reliable networked collaborative office, enhance office information security, improve office efficiency and informatization level, provide information support for scientific decision-making, and promote the leapfrog development of informatization construction.

Abstract： Under the background of building cyberpower, an inherently secure autonomous computing system, i.e., PKS system, comes into being. To promote the continuous upgrades of PKS system, it is necessary to build up the innovation research team. This paper studies the construction of innovation research team in CEC Joint Innovation Research Institute. To tackle with the current challenges of talent work, this paper proposes the guiding ideology and basic principles, four innovative solutions and the corresponding safeguards based on the institute′s 14th Five-Year Plan, which are instructive for its talent work in the 14th Five-Year Plan period.