Skip to main content
main-content
Top

2023 | Book

Signal and Information Processing, Networking and Computers

Proceedings of the 8th International Conference on Signal and Information Processing, Networking and Computers (ICSINC)

Editors: Prof. Jiande Sun, Dr. Yue Wang, Mengyao Huo, Dr. Lexi Xu

Publisher: Springer Nature Singapore

Book Series: Lecture Notes in Electrical Engineering

share
SHARE
insite
SEARCH

About this book

This book collects selected papers from the 8th Conference on Signal and Information Processing, Networking and Computers held in Ji’nan, Shandong, China on September 13-17, 2021. It focuses on the current works of information theory, communication system, computer science, aerospace technologies and big data and other related technologies. Readers from both academia and industry of this field can contribute and find their interests from the book.

Table of Contents

Frontmatter

Space Technology I

Frontmatter
Averaged Equation of Satellite Relative Motion in an Elliptic Orbit

The precision model of relative motion is a necessity for satellite formation flying, but these models are complex for analysis and design, especially in elliptic orbit. For some satellite formation flying applications, the long-term formation maintenance and fuel-saving is more important for satellite life. Using the averaged analysis method over an orbit period, the averaged equation of satellite relative motion in an elliptic orbit is presented in this paper. Firstly, based on the homogeneous solutions of T-H equations, a simple averaged equation of relative motion is derived. Secondly, the improved averaged equation of relative motion which is described by the instantaneous orbit elements difference is developed for considering the orbital perturbation. The effectiveness of the proposed models is verified by four simulation cases which consider the orbit perturbation or not. The proposed model can eliminate the periodicity movement of satellite relative motion, and it is convenient for long-term formation flying designing and configuration controlling.

Rucai Che
HDR Fusion Algorithm Based on PCNN and over Exposure Correction

For traditional HDR fusion algorithm based on mapping functions, the artificial determination of the over exposure region often ignores the spatial distribution and the relation of pixels of the image brightness. In this paper, a stimulated HDR fusion algorithm based on pulse coupled neural network is proposed. For the proposed algorithm, first obtain the external stimuli according to the maximum image brightness, calculate whether the pixel “fires” or not and determine the over exposure region. Then alter the value of external stimuli to perform iteration and construct judgment matrix to make sure that every pixel in the image “fires”. The image irradiance is linear with the illuminance and exposure time. Obtain the brightness mapping function using linear fitting, and synthetize HDR image with over exposure correction algorithm to eliminate artifacts. The algorithm mainly has the following advantages: 1. It fully considers the spatial distribution of the image brightness. 2. For the imaging of different scenes, it self-adapts to separate exposure regions. 3. It performs mapping according to the imaging principles and obtains the mapping function without the exposure time. 4. It reduces the quantitative error of synthetizing image, and the artifacts in the transition region. The algorithm proposed in this paper is experimented under multiple scenes, and the image entropy and spatial frequency are improved greatly.

Fang Dong, Jinyi Guo
A Dynamic Range Analysis of Remote Sensor Based on PcModWin

Space optical remote sensor for earth observation, ground targets and the observation conditions are always changing, as at-sensor radiance changing. Dynamic range is an important factor that can be detected in determining targets, which is an important indicator for measuring imaging quality of space optical remote sensor and radiation properties. The optical sensor design condition is precondition by accurate radiance evaluation on the optical remote sensor. This paper started with an analysis of influence factors for Dynamic Range of The Space Remote Sensor based on radiative transfer and Photoelectric conversion. The Universal calculation of Entrance pupil radiant energy of optical remote sensor was derived from this paper. The calculation provided the basis for the design and evaluation of optical remote sensor. Then, this paper established the radiation transmission model of optical remote sensor for earth observation, using the PcModWin software to calculate each component energy distribution of radiation transfer model, comparing the calculation results with labsphere 1 m integrating sphere radiation data, this method was proved the authenticity and accuracy of the results.

Zhao Ye, Gaochao, Ye Xue, Chao Wang
The Analog Variable Observation Method for S3MPR Controller

MPPT (Maximum power point tracking) is applied in order to reduce the area of solar panels on spacecrafts, improving the power density and reducing the cost. S3MPR (Sequential switching shunt maximum power regulator) combines S3R (sequential switching shunt regulator) power topology and MPPT control strategy and is suitable for high-voltage high-power platform with solar arrays in parallel. In this paper, we propose a new analog MPPT control method (the variable observation method) and use the S3MPR as the power topology. In this new designed method, the dynamic MPPT reference is generated by observing the logic relationship between derivatives of output voltage and power. Finally, a whole analog MPPT control circuit is designed to satisfy the requirements for space applications. The Saber simulation platform is built and simulation results confirm the principles.

Qifeng Xu, Luwei Liao, Junchao Song, Chengrong Cao, Xingjun Lu, Guojun Wang
Analysis and Design of Aerospace Flight Control Knowledge Base

The use of knowledge in the professional field of aerospace flight control to build a knowledge base of aerospace flight control process can improve the execution efficiency of aerospace flight control program and accelerate the progress of automation and intelligence of aerospace flight control system. Propose the representation method of aerospace flight control knowledge and its knowledge base level; use the theory of knowledge classification and first-order logic grammar to standardize the aerospace flight control knowledge; put forward a space flight control knowledge base scheme based on ontology knowledge, and point out its management method.

Zhaoqiang Yang, Jiang Guo, Jian Yang, Ruihua Wang, Daoji Wang, Yang Wu
Study on Planning Strategy of TT&C Resource Scheduling Based on Data Mining

Currently, the scheduling mode used by most TT&C resource scheduling system was based on the mission requirements offered by users. After the scheduling plan is formulated, the resource redundancy has been basically determined, without the advance planning of the resources. This article analyzed the data of the scheduling system, by dividing the requirements of resources into two types of “period” and “peak”. Through analyzing the periodic use of resource historical data, the periodic resource demand will be predicted and guided resources for time-dividing and domain-planning. By analyzing the allocation data of dynamic resource scheduling, the peak resource will be reserved to enable and recycle dynamically in actual. Finally the strategy was verified by simulation, which provided a different approach to improve the utilization rate of the entire network resources.

Ying Shen, Huili Gao, Zhaoqiang Yang, Jianjun Zheng, Wei Zhao
Research on Product Assurance Parallel Work of Spacecraft Control System

This paper mainly studies the practice of product assurance in the process of spacecraft control system development. Product Assurance of spacecraft control system has gone through a series of development processes. Because users have higher requirements for product assurance of spacecraft control system, based on modern quality management theory and combined with the reality of aerospace engineering, this paper presents a parallel research on product assurance based on spacecraft control system. Parallel research method is a systematic method, and its most basic working principle is the working principle of collaborative iteration. In this paper, the product guarantee parallel working model of spacecraft control system is presented, and the product quality of the control system is improved by the parallel research. According to the practice of product assurance parallel work in spacecraft control system, the correctness and validity of the method are confirmed.

Xu Jingyu, Yu Songbai, Wang Hong, Ji Ye
Research on Inter-satellite Ranging Technology Based on Proximity-1 Protocol

The key technologies of satellite constellations and formation flying missions are information interactions, relative ranging/time synchronization between satellites. In this paper, by using the time service given in CCSDS-Proximity-1 protocol, a non-coherent ranging and time synchronization method is proposed to complete the high-precision ranging between satellites while transmitting information between them. The formula derivation of the algorithm and analysis process of the model error are given in detail. The theoretical analysis and simulation results show that the proposed method achieves higher accuracy of ranging/time synchronization and functional integration. It can provide the powerful technical support for the Inter-satellite link.

Wenxia Xu, Lihua Zuo, Jingshi Shen
SpaceWire Network Configuration Method Based on Digraph

SpaceWire is becoming popular as a high-speed (2-400Mbps) and low-cost network. Backup mechanism is often applied in onboard networks to ensure higher reliability. When one node in the network is switched to its backup device, the newly online node needs to be set up as soon as possible to start normal communication. A SpaceWire network configuration method based on digraph is proposed in this article. Network topology is stored in onboard software in form of digraph. The target SpaceWire address from the management node to each router is found by searching path in the digraph so that the configuration of the network can be accomplished automatically. The method helps improve both the reliability and the flexibility of onboard SpaceWire applications.

Yu Junhui, Niu Yuehua, Mu Qiang
LDPC Encoding and Decoding Algorithm of LEO Satellite Communication Based on Big Data

With the rapid development of science and technology today, mankind has entered the era of digitization, and computer and big data technologies have penetrated into all fields and classes of society. This paper first studies two basic decoding message passing mechanisms of LDPC C, and analyzes the performance of the DA through experimental simulation. The application of big data concept in SC network can bring huge performance improvement, and further design the LEO SC network architecture. The results show that the BER performance of the system can be improved by increasing the number of iterations. When the number of iterations is 10, the BER performance is the worst. When the number of iterations is 20, the BER performance is greatly improved. When the number of iterations is 30, the performance is also greatly improved. However, when the number of iterations is 40, the BER performance of the system tends to be stable, and it is difficult to improve the performance of the visible light communication system through iterations.

Nan Ye, Jianfeng Di, Jiapeng Zhang, Rui Zhang, Xiaopei Xu, Ying Yan
LDPC Coding and Decoding Algorithm of Satellite Communication in High Dynamic Environment

In this digital age, modern communication system plays an important role in almost all aspects of life, such as mobile network and satellite communication to the Internet and data transmission. This paper analyzes the concept of high dynamic environment, satellite communication system and LDPC algorithm. The results show that the resource cost of the rate editor of regs is higher than that of LUTS. Look up table LUTS, register regs and two kinds of encoders can be selected.

Xiaoni Wu, Xuexue Qin, Xiaodong Ma, Dayang Zhao, Dan Lin, Yifeng He
Analysis of Optimal Orbital Capture Maneuver Strategy to LEO Spacecraft Based on Space Debris Collision

With the increasing of multi-satellites launching mission, contradiction between operationally orbit maneuver deployment of spacecraft and collision avoidance of space target becomes prominently. In this article, a spacecraft maneuver strategy analysis for the design optimization of orbit capture with collision avoidance is presented. With the linear relationship between maneuvering pulse and spacecraft relative motion is formulated, the analytical model of trace drift nearest distance and crossing space debris time with velocity pulses is proposed. Considering the constraints of TT&C resources, orbit determination accuracy calibration, attitude and orbit control propulsion system calibration, maneuver time limit, space target avoidance and so on, the optimization process of optimal orbital maneuver pulse is designed based on synthetic error estimation. An orbit altitude capture maneuver of triple-LEO Spacecraft launching with one vehicle is designed, the simulation shows that the optimal first maneuver pulse and orbit acquisition strategy are presented, which has good engineering operability and applicability.

Hong Ma, Tao Xi, Shouming Sun, Jianmin Zhou, Qiancheng Wang, Zhenguo Xiao, Wei Zhang
Research on a Centralized Thrust Deployment Method for Space Satellite Swarms

To improve the current deployment method for satellite swarms, this paper put forward a centralized thrust deployment method for space satellite swarms, which is low-cost, high-accuracy, easy to realize and flexible. In this paper, following methods are used: creating “hitch-hiker satellite” swarms combination, satellite swarms combination orbit separation, initializing the mother satellite based on three-axial-stabilization, building geometrical construction and communication strategy of satellite swarms, to satisfy the needs of TT&C and deployment of space satellite swarms, and the application of satellite mission plan and TT&C resource dispatch.

Yuan An, Yaruixi Gao, Xu Yang, Ruolan Zhang, Minzhang Han, Yongjie Zhu
Research on Navigation and Communication Fusion Technology Based on Integrated PNT System

The integrated development of navigation and communication systems has gradually become the mainstream of the development of space-based network information systems. Based on the comprehensive PNT (navigation, positioning, timing) thinking, this article investigates the current status of the integration and development of domestic and foreign communication systems, and summarizes the development status of my country’s communication and navigation systems. On this basis, the system architecture and several methods for the integration of navigation and communication systems are proposed, and combined with the construction of my country’s navigation and communication systems, the main problems and application prospects are analyzed, in order to promote the integration of conduction under the comprehensive PNT system in my country. The application provides development ideas.

Zhao Wenwen, Zhu Xiaofeng, Hu Jintao
A Solar Array on Orbit Output Power Prediction Method for Satellite

An satellite solar array power prediction tool and method is carried out in this paper. The prediction tool including satellite orbit analysis module, the incidence angle of sun light on solar panel analysis model. According to the satellite orbit parameters, the scheme of series-parallel solar cell strings designed, and the solar cell degradation character, we predict the output power of solar wings. The tool can help analyzing power and load balance, mission scheduling and automatic operate management.

Jiang Dongsheng, Peng Mei, Yang Dong, Jing Yuanliang, Du Qing
Conception and Prospects of Exploration and Mining Robots on Lunar Base

Today, the United States, Russia, Europe, and China have all put forward their own plans for lunar bases, and the prelude to the construction of lunar bases as a transit base for deep space exploration has been slowly opened. This article investigates and analyzes the construction plans of various national lunar bases and the results obtained in the exploration and sampling of extraterrestrial celestial bodies. On this basis, an exploration and mining robot for the construction of lunar bases is proposed and briefly explained, include its engineering goals, conceptual assumptions and key technologies, etc.

Zhihui Zhao, Wei Zhu, Ning Xia
Analysis of Orbit Determination Accuracy of GEO Satellite Navigation Data

Aiming at the problem of rank deficiency of orbit determination algorithm equation caused by weak navigation signal, poor geometric structure of signal and small number of visible satellites of a GEO satellite, a weighted least square orbit determination algorithm with additional constraints are proposed based on the relative geostationary characteristics of GEO satellite, which can improve the success rate of orbit determination of GEO satellite navigation signal when the number of navigation satellites are less than 4 and the measurement value is poor. Shorten the initialization time of the receiver. Based on the characteristics of strong radial constraint, weak lateral constraint, strong lateral constraint and strong longitudinal constraint of the ground-based orbit measurement data, the feasibility of the above algorithm is verified by the space-based heterogeneous data fusion orbit determination test. The results show that the accuracy of the two kinds of orbit determination test results are the same, and the root mean square error of the total position in the overlapping trackarc is superior to 5 m. The root mean square error of the total velocity is superior to 2 mm/s. It is concluded that the effectiveness of GNSS system supporting orbit determination of high orbit spacecraft is verified, which provides technical support for the application of GNSS orbit determination mission in the future.

Xiusong Ye, Hong Ma, Yang Yang, Yuancang Cheng, Xing Liu
Variable Bandwidth Digital Channelization Algorithm and Engineering Realization

According to the design requirements of multi-user, multi-rate satellite modems, this paper proposes an adaptive digital channelization synthesis and decomposition algorithm for the number of users and different bandwidth signals. The algorithm is engineered on a hardware platform realization. By adjusting the working frequencies and extracting the results of DFT, variable users and signal decompositions with variable bandwidth can be achieved; or variable users and variable bandwidth signal syntheses can be achieved by adding zeros at the DFT input in practice.

Chen Dan, Wang Xiaotao, Zhang Qi, Shi Min, Wang Xin
A Design Framework for Complex Spacecraft Systems with Integrated Reliability Using MBSE Methodology

In the development of the spacecraft system, reliability analysis is often performed independently of the system design, and its input is usually an artificial abstract of the design scheme, such as function trees or block diagrams. This model gap between analysis and design lead to uncertainty in the results, while the conclusions are difficult to feed directly into the system design. In addition, manual reliability analysis consumes a lot of time and effort, while also increase the risk of errors. With the rapid growth in the size and complexity of spacecraft systems, reliability analysis is becoming increasingly challenging and its value is difficult to show. To cope with these difficulties, model-based systems engineering has emerged. However, MBSE focuses on requirements for traceability without a precise definition of reliability, resulting that the model conversion is still needed. In order to truly exploit the advantages of MBSE, this paper proposes a system design framework integrated reliability analysis for complex spacecraft system using MBSE. Both the design and reliability analysis are based on a single SysML model, from which FMEA and fault tree analysis can be generated directly. This framework ensures the consistency of the system model and the reliability model, eliminating the model conversion and reducing the workload of manual analysis. In this paper, the application process of a simple case is also demonstrated.

Jingyi Chong, Haocheng Zhou, Min Wang, Yujun Chen
Intelligent Task Management of Infrared Remote Sensing Satellites Based on Onboard Processing Feedback

In order to meet the needs of real-time abnormal events early warning missions of infrared remote sensing satellites with multiple observation elements, an intelligent task management method for infrared remote sensing satellites based on onboard processing feedback is designed. Relying on the feedback suggestions of the onboard processing of the satellite-borne infrared images, the users’ observation requirements, satellite design constraints, target area attributes, ground receiving capabilities and other information are collaboratively used to complete the autonomous mission planning and autonomous health management of the infrared remote sensing satellites. On the basis of application and research verification on-orbit, recommendations for mission management under the background of future remote sensing satellite network are given.

Xinyu Yao, Yong Yuan, Fan Mo, Xinwei Zhang, Shasha Zhang
Fragmentation Technology of Spacecraft TT&C Resources Under Multi-constraint Condition

With the development of the world space industry, the number of spacecrafts in the future will increase much faster than the ground-based TT&C equipment and space-based TT&C capacity. In the case of SpaceX's Starlink Constellation program, all 4409 satellites are expected to function by 2024, eventually forming a supergiant Internet constellation of about 42,000 satellites. Therefore, with the rapid improvement of spacecraft technology and quantity, there is an urgent demand for space TT&C capability. How to speed up the reasonable allocation of TT&C resources, make efficient use of existing TT&C resources, and delve the potential of TT&C resources deployment is a practical and innovative mission. Based on China’s many years of space TT&C practice and the practical problems in ground-based space TT&C capability, this paper puts forward a fragmentation of TT&C resources reorganization algorithm (FRRA) under multi-constraint conditions and studies the technology of deep excavation of the ground TT&C capability of spacecraft. The results of this paper are of practical engineering significance to the existing mode and utilization efficiency of space TT&C in China.

Jun Liu, Jun Wei, Yuan An, Yaruixi Gao, Feng Zhang, Ruolan Zhang, Qifeng Sun
Key Technology Research and On-Orbit Verification of Beidou Short Message Communication System of LEO Satellite

Introduced the typical mission requirements of Beidou short message communication technology for low-orbit remote sensing, studied the key technology of low-orbit satellite short message communication system, and proved the correctness of the designed key technology through the research on the data obtained from the actual test on orbit.

Ren Xiaohang, Shan Baotang, Xu Dinggen, Wang Po, Wang Baobao
Summary of Artificial Neural Network Effectiveness Evaluation Technology

In the field of modern military struggle, research and preparation for war issues are becoming more and more important. The evaluation of the effectiveness of platforms and systems affects the research direction of military issues and even military operations directly or indirectly. However, the development of our high-tech evaluation system is relatively slow, and it has not reached the level of promoting the development of platforms and systems. This article starts with the evaluation of the operational effectiveness of artificial neural networks, the research draws on many materials in related fields, summarizes some methods used in the evaluation, and puts forward some thoughts, in order to provide reference value for the development of subsequent evaluation work.

Zhang Yue, Liang Min
Experiment Study on the Impact Limit of Basalt/Aramid Stuffed Whipple Shields

The impact limit equation is an important basis for risk assessment and the optimal design of stuffed Whipple shields. The internationally commonly used NASA impact limit equations and curves for Nextel/Kevlar stuffed Whipple shields (hereinafter referred to as improved NASA equations) cannot be accurately applied to the protection performance predicts of Chinese basalt/aramid fiber stuffed Whipple shields. This paper studies the hypersonic impact damage characteristics of the basalt/aramid fiber stuffed Whipple shields of the China certain large manned spacecraft. Based on the test data of three types of stuffed Whipple shields, two methods are adopted to modify the parameters of the NASA equations impact limit equation, and it is determined that it is suitable for The impact limit equations and parameters of domestic stuffed Whipple shields. The improved impact limit equations can provide higher prediction accuracy, and provide a reference for the engineering optimization design of shields and the risk assessment of on-orbit missions.

Jiangkai Wu, Runqiang Chi, Guotong Sun, Zengyao Han, Baojun Pang, Shigui Zheng
Application of Software Remote Synchronization Mode in Aerospace Products

As the aerospace industry production and operation scale is constantly expanding, there are new requirements for aerospace software system. It needs to support more work in parallel, collaborative scheduling, information transmission and data analysis. So we must reconstruct the original information system construction. In this paper, I analyzed the interactive mode of heterogeneous database, established a heterogeneous multi subsystem data interaction model, described the remote data synchronization mechanism, and at the end, pointed out the next step of research.

ManLi Li, XiaoHong Liang, XingLong Han
Pointing Precision Modification Method for the Low Orbit Satellite in Ka-Band by the Triaxial Antenna

The beam of the low orbit remote sensing satellite in Ka-band is very narrow, so the ground receiving system needs to have good pointing accuracy, so as to realize the stable acquisition of the system and ensure the quality of received data. By comparing and analyzing the error coefficients of the traditional azimuth-pitching two-axis antenna and the triaxial antenna with the inclined turntable, a triaxial antenna pointing correction model algorithm is proposed to capture the Ka-band of low-orbit remote sensing satellites. In this paper, the error coefficients obtained from the integrated non-tower calibration unit are used to correct the actual pointing, and the feasibility and correctness of the algorithm are verified by observing the level change and pointing error of the antenna pointing radio source.

Fei Gao, Mingnuan Qin, Min Liang, Zhuo Zhang
Theoretical Analysis and Experimental Study on Oxidant Depletion Cutoff Method of Satellite Dual Mode Propulsion System

According to the configuration of a typical satellite dual-mode propulsion system, the oxidant depletion cutoff method based on the abrupt pressure change criterion in the system was introduced, theoretical analysis and hot test based on a real dual-mode propulsion system were conducted. It’s indicated that at the oxidant depletion time, the abrupt pressure change at the pressure measuring point of the propulsion system is distinguishable. It can be used as criterion to confirm the oxidant exhaustion. It costs 2~3 s from oxidant depletion to the thrust loss of the orbit control engine, which is long enough for the judgment of oxidant depletion and the implementation of the programmed cutoff instruction. In this hot test, the response time from oxidant depletion criterion triggered to the orbit control engine cutoff is 1.6 s. It’s indicated the design is reasonable and feasible.

Zhen Lin, Mengjie Wang, Xiangning Li
Parameter Optimization Design and Application Analysis of Water-Based Propulsion System

The water-based propulsion system has the characteristics of green propellant and high specific impulse. The above characteristics make the propulsion system light in weight, high in integration, long in orbit life, easy to achieve on-orbit refueling. Consequently, it has a good space application prospect. Aiming at the mutual restriction between the configuration of water-based propulsion system and the requirements of space missions, a mathematical model for parameter optimization design of water-based propulsion system was established, the matching design of water-based propulsion system under different platform background applications was carried out and corresponding performance was given. Meanwhile, the effect of different design parameters on the design of the water-based propulsion system was carried out. Based on the research results, the space application of the water-based propulsion system was analyzed. Research shows that there is an optimal matching design point for the maximum number of days between missions, and there is an approximately proportional relationship between the weight and the scale of the water-based propulsion system. Therefore, the water-based propulsion system is more suitable for spaces with longer mission intervals and lighter payloads platform.

Wenjuan Yin, Zhen Lin
A New Tracking Measurement Method Based on QPSK Spread Spectrum Communication Signal

Hongyan satellite adopts QPSK spread spectrum communication signal, which can carry out signal enhancement ability verification. The key is to realize accurate tracking and measurement based on communication signal, and then realize the positioning function. However, the I and Q branches of the Hongyan QPSK signal use the same code sequence, which is different from traditional satellite navigation signals and cannot directly use the current satellite navigation signal tracking measurement scheme. Therefore, this paper proposes a tracking measurement scheme based on QPSK spread spectrum communication signals. This method does not strictly require that the carrier reproduced by the receiver is cos and sin. The data symbol estimation method is given based on FFT and m sequence. The periodic I, Q data symbol product is estimated, which prolongs the coherent integration time, realizes signal tracking and measurement, and improves the measurement accuracy.

Zhimei Yang, Hao Tang, Xuan Li, Lixin Zhang
Design and Application of OBDH Subsystem for Gaofen-4 Satellite

Gaofen-4 satellite is the first Chinese GEO optical remote sensing satellite with high resolution. Taking advantages of its permanent position over the equator, Gaofen-4 satellite combines the characteristics of high temporal resolution and high spatial resolution. OBDH subsystem is the important subsystem of the satellite. In addition to the traditional telecommand function and telemetry function, compared with the OBDH subsystem of other remote sensing satellites, Gaofen-4 satellite extends the functionality of autonomous management, adds new autonomous layout function of satellite load mission and optimizes telemetry data transfer modes, as well as the characteristics of universe, high reliability and long-life. This paper describes the design and application of Gaofen-4 satellite OBDH subsystem, focuses on the characteristics depicted on above, and presents the on-orbit verification results.

Liu Xin, Wang Luyuan, Wei Yongquan, Yang Peiyao, Zhao Shengdong
Autonomous Mission Planning Software Design and Testing of the Agile Earth Observation Satellite

At present, the mission planning of agile earth observation satellites is usually carried out by the ground control center, and the planning results are injected into the satellite through telecontrol instructions. Due to the natural defects of ground planning, autonomous mission planning on-board is an effective way to improve the intelligent level and efficiency of agile earth observation satellites. The planning function is changed from ground operation to space operation, and the software operating environment has changed greatly. How to design and test the satellite mission planning software for the space operation environment is the key problem to be solved. In this paper, the key technologies of satellite autonomous mission planning software design and verification are described from three aspects: software requirement analysis, software design and software test or verification. This paper focuses on the design diffrecences between the onboard planning software and ground planning.

Xiutao Fu, Xiaogang Dong, Xiaofeng Li, Fan Xu, Shimin He
Research for Non-cooperative Space Objects Detection Methods Based on Image

This paper takes a research on non-cooperative space objects detection methods based on the image. In order to design a new method suited for optical observation equipment with super-large field of view and high sensitivity, paper investigates existing methods and summaries the development trend from the appearing of object detection till now. According to the research, the main driving force for space objects detection methods is the upgrading of hardware used in observation equipment. Nowadays, methods in this field have gradually gain the unity of high sensitivity, high real-time performance and high accuracy. In the future, with the application of novel algorithms, new methods will achieve fast and accurate detection of darker or even obscured objects.

Yunfan Lei, Hongjun Zhong, Long Wang, Yanpeng Wu
Research on a BSDiff-Based System-Level On-Orbit Maintenance Technology of Spacecraft

To resolve the problem that the traditional on-orbit maintenance technologies will uplink a large number of telecommands, a BSDiff-based system-level on-orbit maintenance technology is proposed in this paper. By applying the BSDiff, the new file will be replaced by the patch file for transmitting to the spacecraft by the uplink channel. After the patch file is received by the spacecraft, the BSPatch algorithm will be applied to combines the old file stored in the storage and the patch file. And finally, the new file will be generated and then stored in the memory. Due to the volume of the patch file generated by this method is reduced by more than 80% compared to the new file, the number of telecommands is effectively decreased by the proposed technology, and it also has better portability and scalability.

Yang Peiyao, Zhang Yahang, Liu Xin, Yang Zhigang, Dong Zhenhui, Zhao Fangyuan
Research on the Risk Identification Index System of Software Project

The risk management is a very essential part in software development process. Comparing to the normal projects, there is much more uncertainty in software projects. The software project urgently needs to strengthen the risk management, which mainly displays in the following two aspects: firstly, the risk management is the internal request in various project groups. As a result of the rapid development in software technology and the widespread application of software system in various social aspects, each kind of risk factors and the risk occurrence have greatly increased, and the loss scale that the risk event created has expanded. All these issues have proposed higher management requests for the project groups. Secondly, the unceasing improvement of risk management method and the successful application on many typical projects have attracted many software project groups to carry on their own risk management. Currently, the risk management has overcome the limitation of traditional methods that only adopt some single measure to deal with risks. While, the current risk management comprehensively utilizes each kind of control risk measures, and improve these methods day by day. However, in software development process, Risk Identification, Risk Assessment are the key elements assuring project success.This paper is designed to reconstruct the risk identification system of software projects from the perspective of nine fields of project management.

Xiaohong Liang, Manli Li, Jingjing Zhao
Research on Time-Frequency Location Algorithm of Launch Vehicle Vibration Data

Clay is a bad capacity of Low-frequency vibration, in order to precisely analyzing under 20 Hz-vibration which affecting cosmonaut’s task. Though researching method of Time-Frequency analyzing based on EMD, build rocket vibration data’s Time-Frequency localization mode and arithmetic with rocket vibration data’s characteristic. By actual data tested, proving method of Time-Frequency localization is correct and effectual. Compared to traditionally Fourier analyze, te method achieve under 20 Hz vibration data’s time-frequency-energy localization.

Bin Che, Wendong Zhong, Jue Wang, Zhenglei Yang, Yangsongyi Su
Study on Disturbing Moment Compensation for the Satellite with Large-Scale Rotating Antennas

In this paper, a method of interference feedforward compensation is proposed for the disturbance of satellites with large rotating antenna such as relay satellites with Large-scale Rotating Antennas. This method utilizes the angular momentum exchange device in advance to pre-compensate In order to reduce the disturbance of the satellite by calculating the coupling angular momentum changes of the satellite caused by the antenna motion at different positions. The simulation results show that the method can effectively reduce the interference of satellite when antennas do large-scale rotating which can make the attitude of the satellite more stable and the directional fluctuation of the platform smaller, which is conducive to the normal work of other satellite loads.

Wei Fan, Yin Zhang, Haibo Zeng
Electromagnetic Interference Posed by Wireless Devices Inside Airproof Cabin of Manned Spacecraft

The coupling mechanism of the electromagnetic interference posed by wireless devices inside the airproof cabin of manned spacecraft is analyzed, and a simulation approach of electromagnetic interference of wireless devices inside the whole metal airproof cabin is presented. By use of structure model of one manned spaceship, we set up an electromagnetic simulation model of manned spacecraft and complete the investigation of electromagnetic circumstance distribution inside airproof cabin of manned spacecraft and the electromagnetic coupling simulation between inner wireless devices and outer antenna.

Yan Liu, Kang Yu, Ruixun Chen
Security Enhancement for Noise Aggregation in DVB-S2 Systems

This paper proposes the application of noise aggregation coding technique in the Second-Generation Digital Video Broadcasting system for satellite broadcasting (DVB-S2) regime to improve the security performance. The DVB-S standard was applied in satellite broadcasting in the past, and DVB-S2 extends the scope of applications to include interactive applications, Digital Satellite News Gathering (DSNG) and other professional services. In addition, as the forward link part of the Digital Video Broadcasting system with return channel via satellite standard (DVB-RCS) system, applied in interactive satellite communications, the security requirements for the transmitted information are gradually increasing. This paper aims to apply an optional module, the noise aggregation coding module to the satellite communication systems. With Automatic Repeat-Request (ARQ) technology, noise aggregation technology uses the original channel noise during communication and improves transmission information security through coding technology. The simulation concludes the false bit rate curves of legitimate and eavesdropping users. When the Eb/N0 is larger than 4 dB, legitimate users can achieve 10–7 false bit rate and eavesdropping users can only achieve 10–4 false bit rate. The proposed noise aggregation coding technique in DVB-S2 can reduce the security risks of anti-eavesdropping and information interception, with the advantage of low cost and high security performance, which can be applied to high-throughput satellites.

Ziqi Zhao, Na Zhou, Hanyu Zheng, Pengfei Qin, Longteng Yi
A Pipelined Arithmetic Architecture for the Motion Planning of Multi-DOF Manipulator

This paper presents a pipelined arithmetic architecture for the motion planning of a multi-DOF manipulator. The architecture starts from the complexity of the motion planning algorithm and the requirement for real-time computing results, which can meet the computing requirement by converting between floating-point and fixed-point formats. For addition, subtraction, multiplication, and division, and square root solving operations, the IEEE754 standard single-precision floating-point number format is used for operations; the calculation of trigonometric functions and inverse trigonometric functions uses a 20-bit precision fixed-point number format for processing. In order to improve the processing speed, a pipelined architecture is utilized to complete parallel calculations, which can generate a calculation result every cycle. The simulation results show that the architecture can be used in a multi-DOF manipulator efficiently.

Yalong Pang, Shuai Jiang, Jiyang Yu, Shenshen Luan
Information Hiding in Space Data Link Protocols Based on Spare Value of Fields in Frame Header

This paper analyzes the space data link protocols, and introduces an information hiding scheme based on manipulating the spare value of fields in transfer frame header. An implementation example of this scheme is presented in this paper. Using the spare value to represent the hidden information will not change the structure and statistical characteristics of transfer frame, which ensures the transparency and robustness of hidden information transmission. This information hiding scheme is suitable for the low speed hidden information transmission requiring high degree of concealment.

Guang Wu, Zhelei Sun

Space Technology II

Frontmatter
Optimal Escape Strategy with a Single Impulse in Space Attack-Defense

While interceptor approached target satellite via the long-range guidance section in the space attack-defense, we studied the optimal escape strategy for target satellite with a single impulse. Firstly, based on the principle of spacecraft orbit dynamics, this paper calculated the trajectory parameters after impulse application. Secondly, taking the interception error as the design index, the paper established an evaluation model on escape effect. Through simulation calculation, we analyzed the escape effect under different initial conditions and different impulse vector. The simulation results showed that there was an optimal impulse direction at the fixed position of the orbit, and the escape effect varied greatly at different positions along the orbit. The research results can be a reference for the design of escape strategy in the space attack-defense. Considering the uncertainty of space attack-defense, we can set up the off-line escape database in advance in engineering practice, in order to conduct the optimal intelligent escape while the satellite being in orbit.

Zhengfei Peng, Yang Guo, Shaobo Wang, Yanhua Tao, Dan Wei
A Method of Using Control Chart to Monitor Software Key Performance Evolution

In order to objectively analyze how the key performance of software evolves during the maintenance period, this paper studies the statistical process control and process capability. Based on the performance data of software use process, a method of monitoring the evolution of software key performance using control chart is proposed. According to the monitoring requirements of software, collect the target performance samples, draw the control chart, judge whether the measured value meets the statistical steady state; according to the abnormal mode and out of control point, analyze the software abnormal and make the maintenance plan; under the condition of the normal distribution of the measured value, synthesize the process capability index and process performance index to judge whether the measured value meets the technical steady state; calculate the unqualified product rate to determine whether the failure rate of software products meets the user's requirements; define the reproducibility of the sample set, and examines the reproducibility from two aspects: the quantity value of reproducibility and the statistical steady state of the mean value of the sample set. The method has been fully verified in a task management system.

Fan Zhang, Wei Guo, Liang Qin, Yanzhao Zhao, Zexi Li
Accurate Pointing Control Algorithm for Space Based Movable Antenna Tracking Ground Station Using GPS

The antenna with polarization multiplexing was used to transmission high speed data on mid-high resolution LEO satellite. The computation precision of antenna pointing and tracing ground station shall be 0.1° (3σ) or less. This paper describes the design of algorithm. With GPS data and attitude on satellite, the difference vector between antenna and ground station is transformed from WGS-84 to antenna coordination through series coordinate transformation. The algorithm must be simplified for less time consuming. Meanwhile, pointing accuracy must be ensured. The assembling error of antenna is took into account to pointing earth station more exactly. The test results confirm that the algorithm meets the requirements for precision pointing and less time consuming.

Yong Zhou, Chao Ma, Xiaona Yuan, Jie Pang, Chao Dong
Telemetry Data Prediction Method Based on Time Series Decomposition

In this paper, a prediction method of satellite telemetry data based on time series decomposition is proposed. The HP (Hodrick Prescott) filtering is introduced to telemetry data analysis and prediction for the first time. The HP filtering method decomposes the time series of telemetry parameters into trend term and fluctuation term, by polynomial fitting and seasonal ARIMA model, the predicted values of the two terms are superimposed to obtain the final combination prediction result. The accuracy and effectiveness of the prediction method are verified through the empirical analysis of the telemetry data of the output current of a satellite solar array and the results show that the relative error is less than 0.67% during the one-month prediction period. This method has significant engineering application value in satellite health assessment, fault diagnosis and early warning.

Zhiqiang Li, Hongfei Li, Xiangyan Zhang, Huadong Tian
A Query Optimization Method for Real-Time Embedded Database Based on Minimum Arborescence

In this paper, a query optimization method based on minimum arborescence for real-time embedded database is proposed to solve the problems of memory limitation and real-time access performance limitation of spaceborne computer system. Based on the characteristics of SQL statements, this method applies the minimum arborescence algorithm to the real-time embedded query processing by relying on the relationship between the attributes of each table during the construction of SQL query. This method can improve the processing speed of multi-table complex query statements, reduce the system memory usage, and optimize query processing in embedded database. The verification on SQLite database shows that this method is an effective query optimization method for embedded database.

Xudong Li, Bo Liu, Jian Xu, Jianyu Yang, Mengdan Cao
Design of Reconfigurable Test System for Information Interface Between Space Station Combination with Multi-aircraft

Aiming at the need for space station assembly flight information engineering functional verification, test system design method is proposed reconfigurable. Platform for high-performance system can be configured to be equipment design, combined with relatively flexible, the software system can be configured using the graphical programming software virtual instrument, comprising on the one hand to support a plurality of interfaces between different aircraft designed by reconfigurable Verification capabilities. On the other hand, in the absence of real aircraft docking, the newly developed aircraft can be effectively verified on the ground. Through simulation test verification, the system can quickly realize the simulation verification of the interface with a variety of manned spacecraft, which makes the test process forward, improves the parallelism of test and development, can effectively cover various failure modes, and the system framework is reliable and significantly reduces cost.

Ying Peng, Jing Xu-zhen, Li Jichuan, Tan Zheng
A Dual Model Control of Sensorless High Frequency Permanent Magnet Synchronous Motor Based on the Space Vector Pulse Width Modulation and Six-Step Mode

To effectively drive the sensorless high frequency permanent magnet synchronous motor (HF-PMSM) within the whole range of speed, the dual mode control of the ten pole pairs HF-PMSM is proposed. Based on the equivalent hardware configuration of the HF-PMSM drive, the typical space vector pulse width modulation for the sinusoidal-wave drive mode is utilized as the startup strategy. The sensorless variable voltage and variable frequency control algorithm is carried out simultaneously. The sensorless six-step square-wave drive mode based on three-phase inverter bridge pulse width modulation is utilized at the high frequency operation. The key rotor commutation position is detected with the phase back electromotive force method. The three phase voltages are obtained by the divider resistances and filter capacitances in the designed zero crossing point detection circuit. Then the commutation signals are detected by the designed detection circuit and software phase shift compensation. The simulation results show that the proposed control method is effective.

Yanzhao He, Zhenyan Wang, Quanwu Wang, Qingrui Bu, Jufeng Dai
Development Mode of Remote Sensing Satellite OBDH Software Based on Prototype Software

In the field of remote sensing satellite, the common framework of OBDH software requirements for each satellite is similar, but there are many subtle differences between each other. Although some reusable components were accumulated in the early stage by means of software components and common requirements design, the overall framework of software and many modules that cannot be fully reused are still difficult to be continuously accumulated and inherited as organizational assets. This paper analyzes the difficulties and problems in the current software development mode, puts forward a remote sensing satellite software development mode based on the prototype software, designs the prototype software to meet the different needs of several satellites, and defines the reuse and improvement principles. The application in several satellites shows that this mode can effectively improve the development efficiency of common function software in remote sensing field.

Dong Zhenhui, Zhang Yahang, Wang Xianghui, Yang Peiyao, Zhang Hongjun, Yuan Yong, Liu Yiming
Design of Closed-Loop Tracking Mode for Remote Sensing Satellite Data Transmission Antenna

The remote sensing satellite data transmission antenna is used to track the ground receiving station in orbit, and transmit the radio frequency signal with load data down. The tracking accuracy of the data transmission antenna to the ground receiving station directly affects the smooth execution of the satellite mission. The imaging mission characteristics of remote sensing satellites are analyzed, and the closed-loop tracking mode of the data antenna based on angle and angular acceleration is designed. The data antenna controller receives the data of the antenna pointing angle from the control computer, and calculates the angular velocity of the antenna to track the ground station according to the predetermined control algorithm, uses the angular velocity to control the antenna to track the ground station continuously, and feeds back the position angle and angular velocity of the antenna to the control computer periodically. The results of ground simulation shows that the tracking accuracy of the data transmission antenna using the closed-loop tracking mode meets the requirements of remote sensing satellites, laying a foundation for the design of inter-satellite data transmission links in future.

Yong Yuan, Yufei Huang, Xinyu Yao, Xinwei Zhang, Zhenhui Dong
High Accuracy Orbit Control Strategy of Change’4-Relay Satellite

In order to ensure the long-term stable operation of the satellite in orbit, the Change’4-relay satellite has put forward a very high requirement for the orbit control accuracy. Aiming at the incremental error around thrust direction, quantization error of sampling interval, thrust error of engine and zero deviation of accelerometer, this paper formulated the installation layout optimization of thruster, sampling quantization error suppression, thruster after-effect suppression and zero deviation suppression of accelerometer respectively. The in-orbit control results show that the control strategy makes the orbit control accuracy exceed the target requirements and extends the satellite’s in-orbit life simultaneously.

Tao Zhang, Rongxiang Cao, Jianmin Zhou, Jianzhao Ding, Yifeng Guan, Weiguang Liang
Observation Data Quality Evaluation of GPS Receiver on Agile Platform Based on Ground Simulation

In this paper, the observation data quality of GNSS receiver on agile platform is analyzed and evaluated by the ground semi physical simulation method, which has important reference significance for analyzing the navigation data measured in orbit after the launch of agile platform mission. According to three different observation conditions, the simulation of observation value is carried out by using Spirent signal simulator. According to the characteristics of the simulation observations, this paper proposes to use the observation combination without geometric distance to analyze the simulation pseudo range and phase observation accuracy, and according to this, three groups of different simulation data are processed and evaluated. In addition, the geometry orbit determination is carried out by using pseudo range and phase observations respectively, and the accuracy level of the orbit determination is analyzed. The accuracy of the observations is analyzed based on the residual of the observations.

Jinghua Wang, Xingchen Liu, Ning Liu, Wenwen Li, Yilong Bai
A Station Guidance Method Based on Combination Deviation Spacecraft Orbit Prediction

In order to complete the space-ground coordination tasks during the autonomous control of the spacecraft, the ground-based measurement and control system needs to simulate the autonomous control process to predict the orbit of the spacecraft and guide the station to track it. In general missions, the spacecraft’s orbit forecast ephemeris is used to calculate the epoch and azimuth angle corresponding to a fixed elevation angle under the station coordinate system, and use this as the guidance information. For autonomously controlled spacecraft, the ground system cannot accurately simulate the motion process of the spacecraft and predict the orbit. When the predicted orbit deviation is large, the use of the above method may cause the equipment to fail to capture. In order to solve the problem, a spacecraft station guidance calculation method based on combined deviation orbit prediction is proposed. When the station uses the guidance data calculated by this method to capture, it only needs to wait at the theoretical waiting point, and the waiting time is a certain interval. The method makes the station operation simple and easy to judge the tracking situation.

He Yufan, Huang Jingqi, Li Junfeng, Zhao Ning
A Distributed Real-Time Orbit Determination Algorithm Based on Asynchronous Multi-sensor

To solve the problem of asynchronous sampling and communication delay of sensor network in Real-time Orbit Determination, an asynchronous distributed algorithm based on information filtering is proposed. local state information and measurement information with sampling time are transmitted between local sensor and adjacent nodes in a certain topology structure. The local sensor sorts the received asynchronous information by time, and calculates the target state respectively. This method is simple to implement and the frequency of communication between sensors is small. Algorithm has been checked by real measured data of the first BDS-3 GEO satellite, the results show that the orbit determination results of each station converges to a centralized estimate, and the precision of 100 s real-time orbit determination results can be used to quickly evaluate the control effect.

Shanpeng Sun, Jingqi Huang, Jue Wang, Junying Hong, Yangsongyi Su
An Equivalent Calculation Method of VLBI Instantaneous Time Delay Rate

Radial velocity needs to be introduced into the instantaneous time delay rate model of VLBI, which will increase the partial derivative calculation of the velocity term in the observation matrix, and the state transition matrix will also have an extra state transition part of the velocity term. The calculation amount of the state transition matrix accounts for a major proportion in the precise orbit determination, so it is of great significance to reduce its scale for improving the operating efficiency of orbit determination software. An instantaneous delay rate equivalent calculation method is proposed, which requires only the position and avoids the velocity term and its partial derivative. Compared with the instantaneous time delay rate model, this method reduces the state transition matrix from 6 × 6 dimensions to 3 × 6 dimensions, and reduces the calculation amount of software while ensuring the accuracy. The measured data of the Earth-Mars transfer segment of “Tianwen-1” is used to verify the effectiveness and accuracy of the method.

Huang Jingqi, Wang Fan, Ye Nan, Wang Yanrong, Sun Shanpeng
Design and Research of Satellite Fault Real-Time Simulation and Verification Platform Based on FMI Standard Model

In order to solve the problem of “new design, many states” and time-consuming on satellite fault verification, the paper proposes a design method of real-time simulation and verification of satellite fault based on FMI standard model. Firstly, based on the model design of the general FMI standard, a set of general satellite model library is built; Secondly, based on FMEA automatic fault management design, the fault classification verification and closed loop management is achieved; Finally, based on the fault injection design of the model, a real-time fault simulation verification platform is built to realize the simulation verification of different types of faults on the satellite system level, at the same time, it has the functions of state monitoring, process control and verification management in the process of fault verification. Through the research in the paper, the satellite comprehensive test can be effectively supported, and the satellite in orbit stable operation can be guaranteed.

Liu Jian, Liu He, Wang Huamao
A Method for Calculating Solar Pressure of TDRS with a Movable IOLA

In this paper, the solar pressure model of TDRS (tracking and data relay satellite) with a IOLA (inter-orbit link antenna) of large movable is studied, according to the actual satellite model, it is simplified and discretized; The geometric relationship among mesh antenna, satellite body and solar wing are accurately described by the projection calculation of the rotation of user satellite tracked by IOLA when the user satellite is tracked by antenna; By projecting the information of each part of the discrete satellite along the vector normal plane of the sun light, the shelter relation and the accurate solar pressure area can be obtained. Through the above method, the calculation of the solar pressure of the TDRS with a IOLA of movable mesh tracking the user satellite can be effectively solved.

Yin Zhang, Dong Han
Research on Platform Construction of Small Satellite GNC System

A diversified development trend is occurring in the field of payload missions of small satellites. On the basis of traditional remote sensing, communication, and navigation, applications such as constellation networking, space manipulation, and inter-satellite collaboration have been expanded. High-performance satellite platforms are required to provide supports for these missions, especially for satellite control systems that require high-precision, high-stability, and high-sensitivity, as well as the ability to quickly complete integrated development. This article puts forward the viewpoint of efficient design and development of small satellites based on generalized platforms. Taking the platform-based construction of the GNC system as an example, a platform-based design scheme was proposed for function modularization, interface standardization, hierarchical software structure, configuration of general requirements, customization of special requirements, and automation of testing and verification.

Xin Zhang, Qingyuan Tang
The Storage Optimization of Test Cases and Test Results in Spacecraft Automatic Test Based on JSON

In the face of the complex requirments of spacecraft efficient automatic test in intelligent autonomous mission, a storage optimization method of spacecraft test cases and test results based on the JSON is proposed in this paper by using the idea of combining relational database and no relational database, which effectively stores the automatic test cases and test results of different kinds of satellites in a unified database. Experiments show that this method saves the storage space of test cases and test results, and shortens the data retrieval time. It solves the requirments of modern spacecraft intelligent test for flexible format, easy to modify and fast rendering speed of automatic test cases and test results. This method greatly improved the automatic test capability of spacecraft. It plays an important role in completing the complex spacecraft automatic test task.

Hongjiang Song, Haiyang Chu, Xin Wen, Wei Lv, Xiaoyu He, Haixiang Zhang, He Gao
Research on the Modeling and Application of Intelligent Factory Facing the Complex Space Optical System

Recent years, the application requirements on civil and commercial remote sensing areas are gradually growing. As the most import equipment of space optical remote sensor, the complex space optical system needs to adopting intelligent manufacturing mode for improving its whole manufacturing capability and solving the problems of long manufacturing cycle and low-quality consistency. To meet the requirements, based on the research status of enterprise modeling and intelligent factory modeling method for the optical system, a multi-dimension intelligent factory modeling framework for the complex space optical system was proposed. It contained three dimensions: 1) Business hierarchy dimension based on intelligent manufacturing technology; 2) Development lifecycle dimension based on product digital twin model; 3) Knowledge dimension for crossing disciplinary and domains. This modeling method has features of integration, reusability, multi-level and extendibility. And it was applied on the construction process of the intelligent factory for complex space optical system. Therefore, it is expected to be widely used in intelligent factory construction and implementation on space optical system manufacturing field.

Ying Che, Hui-li Fan, Feng Guo, Yong-sheng Pei, Li-ling Liu
A New Space-Based Optical Surveillance System

In order to monitor the large deployable equipment in space, optical surveillance system is introduced. Applying the system, We can not only monitor deployment status in real-time, but also obtain the dynamic characteristics of the deployment process. It can also Provide necessary information for large-scale deployable flexible mechanism design, abnormal discovery and analysis, etc. First, on the basis of analyzing the characteristics of the system’s tasks, the surveillance system solutions are introduced, including camera solutions, video transmission solutions, image processing, and radio frequency modulation solutions. Second, the moving target image detection and recognition technology and the development of dynamic analysis technology are studied. Finally, the key technologies were tested and the test results were analyzed and discussed. The test results prove that the system can effectively complete the monitoring tasks of large-scale deployable equipment.

Yang Sun, Ling-ling Chen, Xuan Li, Ye Yang, Ling Liu
The Application of MPLS VPN in Improving the Reliability of Test Network

At present, the data transmission of the test network is interrupted due to the breaking of optical cable or the failure of transmission equipment. In order to ensure the reliable transmission of the test network data network and avoid the problem of data transmission interruption in the implementation of the task, redundant lines and equipment need to be added. In this paper, without adding the current test network equipment and links, make full use of the existing equipment idle interfaces and links, using MPLS VPN technology, virtual backup router, in order to achieve the hot backup of the main equipment. Thus, the data of the test network can be transmitted more reliably and the network failure recovery time can be shortened. Due to the support of multi-protocol, MPLS carries a very rich type of business, which enables it to bring new applications in new fields, such as it is used to solve the problem of intranet interworking between companies and enterprises.

Hui Wang, Yuhua Cao, Quanwu Wang
Performance Analysis of the Photon-Counting LIDAR Based on the Statistical Property

Combined with Lidar equation and single photon detection probability theory, the photon-counting Lidar performance indexes under different parameters are studied. The analysis shows that when the surface reflectance is relatively low and the slope is gentle, higher detection probability can be caculated by increasing the grid size of the ground surface. In the case of 1 MHz noise rate, when the frame threshold of the photon counts is 3pe, the detection probability of 99.78% can be achieved. In the case of 10 MHz noise rate, when the frame threshold of the photon counts is 7pe, detection probability of 97.7% can be achieved. Finally, a feasible parameter suggestion of photon-counting spaceborne Lidar is given.

Jun Dai, Shaohui Li, Fei Gao, Haiyi Cao, Gaofeng Guo, Xigang Liu, Qianying Wang
A Space Vector Transformation Based Dynamic Image Rotation Compensation Model

In the field of space-borne laser communications, the acquisition, tracking and pointing (ATP) system plays a key role to guarantee a stable, reliable, persistent inter-satellite chain. However, suffering from the limitation of the assembly mechanism and the specific structure of the optical path, the image rotation and the optical pointing error are inevitable, which deeply influence the chain-establishment time and tracking accuracy. In this paper, by analyzing the vector optical transmission characteristics of the periscope-type coarse pointing equipment, a dynamic image rotation compensation model is proposed. Using the so-called vector compensation matrix, the image rotation and the optical pointing error could be well overcome. Simulations demonstrate that one could obtain a high-approximate estimation of the optical transmission model at any position.

Bin Ren, Qingkun Xie, Jingying Bian, Qian Lu, Xiaoqi Liu, Le Zhang
Research on Segmented Weighted Multi-vector Attitude Determination Method Based on On-Orbit Accuracy of Star Sensors

A segmented weighted multi-vector attitude determination method based on on-orbit accuracy of star sensors is studied to improve the fact that the current attitude determination methods cannot make full use of the measurement data when a large number of star sensors are equipped on a satellite and the problem that the attitude determination accuracy is susceptible to the temporary decline of the measurement accuracy of one single star sensor. The proposed method is based on QUEST algorithm, where the weighted coefficients are meliorated considering the layout angles of star sensors, mechanical installation structure and on-orbit measurement accuracy of star sensors. A mathematical simulation using on-orbit data shows that the attitude determination accuracy can be maintained at the high level, when the measurement accuracy of a star sensor declines.

Wang Jiawei, Jing Quan, Li Shaohui, Gao Hongtao, Mo Fan, Chen Xi
The Research of Micro-vibration Nonlinear Characteristic Produced by Spacecraft Mechanical Momentum Wheel

Micro-vibration produced by momentum wheel is the main disturbance source for spacecraft, with obvious nonlinear characteristics. The researching and general situation of spacecraft micro-vibration produced by momentum wheel, includes vibration sources mechanism, dynamic model and vibration control methods, are briefly reviewed in this paper. The nonlinear characteristic, frequency and amplitude of micro vibration, are analyzed respectively. The momentum wheel’s nonlinearity is shown in two respects: the nonlinearity of amplitude and the nonlinearity of frequency. The movement of moving parts is the fundamental cause for micro-vibration and nonlinearity. The main source of frequency non-linearity is the rolling bearing. Passive or active vibration control method which is essentially using dampers or piezoelectric actuators to realize vibration isolation. The paper discusses the researching contents and direction of these problems mentioned. In the end, some comments are made about the future development in this field.

Quanwu Wang, Yiping Yao, Zekun Yang, Xiufeng Zhong
Design of On-Orbit Monitoring Method for Small Satellite Payload Equipment Based on Grey Prediction

The realization of the on-orbit function of small satellites mainly depends on the normal operation of payload equipment. At present, the number of small satellites in orbit is increasing, same as the tasks of the on-orbit management are increasing. At the same time, many factors, such as the large differences between payload equipment, the high proportion of overseas work, and the working environment of short-term power on orbit, are not conducive to the on-orbit health monitoring of payload equipment. In order to find out the potential faults of payload equipment in time, a kind of on-orbit monitoring method based on grey prediction is proposed. The method optimizes the normal monitoring method and makes grey prediction analysis after data processing of the original telemetry information. At the same time, the monitoring criteria are adjusted autonomously according to the analysis results, so as to achieve early warning and reduce the operation process of ground personnel. High precision and high feasibility of the grey prediction model are verified through simulation by selecting the telemetry information of a satellite. This monitoring method can be further extended to the parameter monitoring of other single equipment and the applications of other types of satellite missions.

Yandong Han, Jian Shi, Guorui Yan, Da Lv, Lei Ma
Investigation on EMC Design of Spacecraft Under High-Power Microwave

This paper speculates the HPM effect on the spacecraft and the corresponding suggestion of design. The HPM energy enters into the spacecraft through two paths, front-door coupling and back-door coupling. The mitigation suggestion for front-door coupling is microwave limiter, FSS, ESS, and EFSS. HPM effect through back-door coupling is simulated. Cable penetration influences the SE a lot. Except the metal honeycomb structure with multi-layer blanketed, extra shielding woks for the penetrated cables with aluminized layers are suggested. After shielding, the protection requirement of 30 dB is proposed.

Ran Li, Dianjun Wang, Hanlu
Ground Experimental Verification of High-Stability Temperature Control System and Flight Data Analysis

TianQin project initiated China gravitational waves in space. Base on TianQin-1 satellite, the main payloads had required a temperature stability environment specifically better than  ±0.05 K. High-accuracy temperature measurement (HATM) is the basis of High-stability temperature control (HSTC), as well as the ground experimental verification (GEV) is the basic of HATM system. The HATM of GEV of TianQin-1 satellite and Flight Data Analysis had been described and discussed in this paper.

Ran Wei, Yupeng Zhou, Yelong Tong, Qing Yang, Ming Li, Lihua Zhang, Xin Zhao
Experimental Verification of Discharge of Two Parallel Sets of Batteries in Mars Probe During EDL

Mars exploration missions have the characteristics of diversification and complex environment, realize Mars circumnavigation, landing, and patrol, and carry out Mars global and comprehensive circumnavigation detection and regional patrol detection. Based on the complex characteristics of tight timing in the entry, descent, and landing (EDL) segment, irreversible events, non-interference from the ground, and large time delays, this paper proposes an architecture and strategy of power system. During the EDL phase of the probe, the Mars rover lithium-ion battery pack is used as a backup for the lithium fluorocarbon battery pack of entry stage for supplementary power supply. The two sets of battery packs parallel discharge to complete the task of entering and landing. The strategy is verified by discharge experiment, and the effectiveness of the parallel discharge is further obtained by test data of the two sets. The proposed strategy is to ensure the reliability and safety of the power system under the complex working conditions of the EDL section.

Jing Wang, Haijin Li, Yihong Liu, Xinjun Liu, Hao Mu, Yi Yang, Dong Yang, Xuerong Qiao, Yao Zhou
Reliability Design and Experimental Verification of Impact Sensor Signal Cable

Deep space exploration mission has the characteristics of diversity and environment complexity. More stringent requirements are put forward to impact monitoring capability when detector enters and lands target planet. Through cables impact sensor transmits impact signal to signal acquisition and processing unit. In order to ensure impact signal channels reliability, this paper proposes a design method of the printed circuit board transfer unit (PCB transfer unit) to connect impact sensor and detector cables, to solve the problem that the connection between the impact sensor and the cables of Mars detector does not fulfill the requirements of spacecraft welding process. In the meantime, on-orbit space environment adaptability and signal transmission reliability of the PCB transfer unit are verified through ground appraisal mechanical and thermal environment tests. The reliability design and experimental verification of impact sensor signal channels are important for the smooth realization of space exploration mission.

Yi Yang, Yuqi Qin, Jing Wang, Bingxin Zhao, Yihong Liu, Yue Han
Simulation and Analysis of Influence of Group Delay Distortion on Satellite-to-Ground and Satellite-to-Satellite Link Channel

In the high-speed satellite-to-ground and satellite-to-satellite data transmission, the signal suffers significant group delay distortion because the bandwidth is limited. Simultaneity, as high-order modulation has high spectral efficiency, it will be widely used in the high-speed satellite-based data transmission. At the same time, High-order modulation is sensitive to the group delay distortion. An 8PSK system is simulated to analyze the influence of group delay distortion. The results shows that when the group delay distortion is 4 symbol periods, the performance of 8PSK modulations is acceptable.

Rui Wang, Yu Wang, Xige Hu
The Influence of Ka Band Satellite Transponder Nonlinearity on High-Speed Data Communication

Satellite communication requires higher and higher information transmission rate, which also requires the development of satellite transponder from traditional C and Ku frequency band to a higher Ka frequency band. However, the performance of high-speed data transmission will be seriously affected by the nonlinear distortion of transponder. In this paper, an actual Ka transponder is specially constructed, which uses a Cortex demodulator to send 600 Mbps and 747.5 Mbps high-speed data transmission signals respectively, and receives and demodulates the high-speed signals transmitted by the transponder, to study the influence of the transponder non-linearity on the bit error rate. Through test, it is proved that the non-linearity has a great effect on the high-speed data transmission system. It is also proof that in the case of high-speed data transmission the performance is not better when the input backoff is more. An ideal input backoff value is 3 dB for Ka transponder in the test. Through the test of Ka transponder in practice, this paper provides a practical reference for the selection of the backoff value in the application of Ka transponder at high-speed data communication.

Zhiyu Zhou, Junyi Zhao, Baoxiang Song, Bingzhe He
GPS and Laser Altimeter Data Alignment for Snow Roughness Measurement in Antarctic

The snow roughness at centimeter scale is an important parameter in the study of Antarctic ice cap. It is also an important parameter for the analysis of radar altimetry and SAR imagery over the Antarctic. The work done in this paper is to increase the resolution and accuracy of the roughness measurement by resampling and aligning the laser altimetry data to the GPS time grid and by using different frequency filter and correlation methods.

Sheng Zhang
Angles-Only Onboard Initial Relative Orbit Determination with Auxiliary Satellite

Onboard initial relative orbit determination (IROD) by using angle measurements is the key to Space Situational Awareness. The camera offset measurement method is one of the most used IROD methods when only the angle measurement is available, which can overcome the range-observability problem caused by signal-camera measurement. However, the camera offset method cannot estimate the relative-position/-velocity when the chaser-target distance is much larger than the camera offset value, which is often below five meters. To overcome this problem, an auxiliary satellite treated as the chaser’s camera with known offset is proposed in this paper, which is suitable for satellites without cameras, and the IROD algorithm is also conducted correspondingly. The Tschauner–Hempel relative orbital motion dynamics and measurement model are given first. Then the IROD algorithm is yielded as a compact form by using the state augmentation least square method. Finally, a two-body satellite’s orbit dynamics model is built to generate simulation data, and three scenarios are studied to verify the convergence of the proposed IROD algorithm. Simulation results validate the proposed IROD algorithm converges quickly, and the measurement time corresponding to the lowest relative-position estimation error is related to the distance among the target, the chaser and the auxiliary satellite.

Jiaxing Li, Li Yuan, Cong Zhang, Sihang Zhang
Space-ROS: A Novel Real-Time Operating System Architecture and Implementation for Space Robot

Robot operating system has become an important basic software for robots. It improves the reuse rate of software codes and provides a unified platform for application developers. However, ROS runs on Linux, which takes up a large amount of memory and cannot provide timing guarantees for robot motion. Although the strategy of porting ROS to embedded real-time operating systems has become a popular solution to these problems, this technique is challenging to efficiently adopt in space robot systems due to real-time multi-tasking, scalability, and performance requirements. This paper proposes an operating system architecture Space-ROS based on the embedded real-time operating system SpaceOS, we introduce a part of ROS function instead of overall on RTOS to ensure the lightweight of the system. Space-ROS adopts a three-level scheduling strategy suitable for multi-cores processors to ensure the efficient operation of real-time multi-tasks and uses a dynamic loading function to improve the scalability of the system, this function supports online updates of applications. In addition, we have also designed intelligent interactive functions, which can conveniently plan and design the action tasks of space robots, greatly improves the efficiency and interactive experience of developers. Finally, we built a simulation environment for the space robot performing a lunar exploration mission and evaluated the performances functions of Space-ROS.

Zhi Ma, Anhua Deng, Jingjing Jiang, Hongbiao Liu, Xi Chen
The Conception of the Construction of Remote Sensing Satellites Information Quality Evaluation Test Field

The information obtained by remote sensing satellites is a major source of important intelligence, and the correct evaluation of the quality of this information can promote the development of related industries. Through literature review and investigation of typical units, the targeted demands of remote sensing information users, such as front-line users, space launch site and camouflage technology materials, were analyzed and summarized. Concerning the technical indicators of remote sensing information quality evaluation at home and abroad, it is planned to carry out personalized and objective full-reference, partial-reference and non-reference evaluation research for user needs. The ideas of sub-regional construction, step-by-step demonstration and multi-unit cooperation are given.

Dalei Luo, Fengtian Bai, Chenming Bao, Qian Liu, Chongying Lu
Design and Optimization of Permanent Magnet Brushless DC Motor Control System for Flywheel

This paper analyzes the working characteristics of the reaction flywheel commonly used in satellite attitude and orbit control system, and selects permanent magnet brushless DC motor (BLDCM). Firstly, the BLDCM working principle is introduced and the mathematical model is established. Secondly, the double closed-loop speed control system of flywheel is established according to its mathematical model. Then the design of the speed controller control algorithm is optimized, and using the adaptive PID control meets its wide-speed range operation requirements and improves the debugging efficiency. Finally, the feasibility of the algorithm is verified through experiments to meet the design requirements.

Chenming Bao, Bin Li, Ailin Kong, Dalei Luo
On-orbit Research and Validation on the Lifetime Capability of Propulsion System

Nonmetallic material is widely used in the satellites’ propulsion systems. In order to analysis the lifetime of the propulsion systems, the compatibility especially the rubber material compatibility should be test. A method for researching the rubber compatibility of a propulsion system in orbit is proposed based on the fact that the compatibility test is carried out only under the standard condition nowadays. In this method, the conventional method on the standard condition is consulted and the system condition variety by catalytic decomposition under pressurized, high-pressure condition is considered to get the exact catalytic decomposition rate. At the same time, the rubber compatibility is performed based on the system pressure variety. According to the rubber compatibility research, we can know that the catalytic decomposition rate of the rubber at high pressure is obviously lower than that at standard pressure, which ensures the margin of the test result aground applied in orbit. At the same time, some in-flight orbital controlling test is carried out to validate the lifetime of propulsion system. And some suggestion of rubber application in propulsion system is proposed according to the catalytic decomposition rate and the system condition in orbit.

Jialong Ji, Yuan Wang, Xin Miao, Tao Chen
Observer-Based Variable Structure Control for Spacecraft with Actuator Failure

Aimed at the problem of actuator efficiency loss failure of spacecraft attitude control system, a variable structure fault-tolerant control method under the influence of interference is studied. Firstly, a sliding mode fault estimation observer is designed to realize effective estimation of actuator failure. Secondly, using the output of the observer and considering the suppression of external interference, an adaptive sliding mode control method is introduced to estimate the unknown parameters in real-time. Finally, based on Lyapunov theory to prove the stability of the controller, the simulation results show that the method of execution efficiency loss spacecraft failure has strong fault tolerance.

Haibin Qin, Jianxiang Xi, Jun Liu, Xu Yang, Nengjian Tai
Research on the Method of Satellite Energy Real-Time Dispatching

The satellite energy management is great significance in dealing with power supply and distribution system failures and ensuring the reliable operation of spacecraft. The article proposes a real-time energy dispatching method on the satellite, which can judge the power supply capacity and load status of the energy system on the satellite, and accurately calculate the energy status, avoid the calculation deviation caused by the inaccurate collection of ground budget parameters. Autonomous calculation of on-board energy status in multi-change working mode and multiple combinations greatly improves on-board energy utilization and enhances the satellite’s business capabilities.

Ming Qiao, Xiao Fei Li
System Error Registration Method of Bearings-Only Passive Location System for Surface Ships

Aiming at the problem of system error registration in the bearings-only location system for surface ships, this paper proposes a registration method based on the genetic algorithm (GA). The presented method directly uses the pure angle observation information of multiple passive sensors to construct the optimal objective function according to the basic principle of error registration. The original problem is then transformed into a nonlinear parameter optimization model with constraints, and GA is used to solve it. The coding mode, fitness function, selection, crossover, mutation and other iterative processes of the GA are properly designed. Numerical simulation results demonstrate that the proposed method can achieve the error registration effectively with high accuracy and reliability.

Zhi Guo, Chunyun Dong, Zhenyong Bo, Xu Yang, Yang Yang, Tao Xi, Jian Zhang
Research on Passive Location of Maneuvering Target Based on IMM-SRCKF Algorithm

To address the problem of three-dimensional non-cooperative maneuvering target tracking in the air, a multi-sensor cooperative passive positioning solution based on the Kalman filter framework is proposed. Firstly, the constant velocity (CV) model and the Singer model are involved to approximate the unknown kinetic characteristic of the non-cooperative target. Secondly, the network positioning is carried out by multiple passive sensors deployed in different areas on the ground, and a centralized observation equation is established. Then, the registration error of each sensor is extended to the target state vector considering its influence on the positioning accuracy, and a unified state space model is built up. Finally, the interacting multiple model (IMM) algorithm based on the square root cubature Kalman filter (SRCKF) filter is devised to realize the estimation of the target state as well as the registration error. Numerical simulations indicate that the presented algorithm can achieve the joint estimation of the target state and the sensor registration error in real time with high accuracy.

Chunyun Dong, Zhi Guo, Xiaolong Chen, Xu Yang, Yang Yang
A Standard Driver Design of Onboard Control Computer

Driver control is a unique program for communication between computers and equipment, an integral part of software development. In the traditional satellite design, each computer product has its functions; thus, the supporting software also needs to be developed separately, which leads to the heavy workload of software repeated development. At present, the functions of the onboard control computer are becoming more and more complex. It is hard to reuse the software, which increases the system risk. This paper establishes a standard software driver function library and integrates anti-SEU design and configurable design to further enhance the reliability of software in-orbit operation. At the same time, it can realize rapid configuration for different products and improve the reuse rate of software. At present, the method has been successfully applied in scientific research, which achieves a good effect.

Chenlu Liu, Jian Xu, Zhi Ma, Fei Peng, Mengdan Cao, Jianyu Yang
Optimal Multiple-Impulse Noncoplanar Rendezvous Trajectory Planning Considering the J2 Perturbation Effects

The minimum-fuel optimal multiple-impulse noncoplanar rendezvous trajectory planning considering the J2 perturbation effects is mainly studied. The multi-impulse orbit dynamic model is constructed based on the dynamic equation considering J2 perturbation and the theory of impulse orbit maneuver. To automatically satisfy the terminal condition,the Lambert algorithm considering J2 perturbation is introduced to solve the last two impulses. Taking the impulse time and delta velocity as optimization variables, and the total velocity increment as optimization objective function, a multi-impulse orbital rendezvous planning model was established.The optimization solution is carried out by particle swarm optimization algorithm (PSO). The simulation results show that the proposed method is effective by using three impulses and four impulses orbit maneuver to achieve the rendezvous with noncoplanar targets.

Guangde Xu, Zongbo He, Huiguang Zhao, Qiang Zhang, Yanjun Xing, Cong Feng
PIC-MCC Simulation of the Temporal Characteristics of the Plasma in a Hall Thruster

In order to investigate the temporal characteristics of the plasma in a Hall thruster with a magnetically shielded configuration, a 2-D cylindrical PIC-MCC model of the Hall thruster is established. The plasma parameters and the sputtering erosion related parameters are derived and obtained. The results indicate the designed combination of the magnetic field and the channel walls remains relatively excellent propulsion performance while achieves near-zero erosion of the channel walls when the anode voltage is 350 V. The results can be utilized for the design improvement of new Hall thruster with optimized propulsion performance and prolonged lifetime.

Rui Chen, Li Wang, Xingyue Duan
Research on Automatic Measurement of Aerospace Electrical Connector Wire

Aerospace cable network products have the characteristics of large cable network, difficult transportation, many kinds and types of cable plugs, small batch, complex cable signal connection, etc. in the early stage, the loose wire labeling of low-frequency cable connectors mainly relies on manual operation, using common tools such as conductors and multi-meter for inspection, and the test operator generally needs at least two or more people, According to the design drawings and the cable wiring table, the test is carried out one by one. The test time is long, and it is easy to make people tired during the inspection process, resulting in cable detection errors. Therefore, the loose wire conduction number of aerospace cable connector is a major reason for hindering the production efficiency of cable network, This paper designs the scheme of automatic number measurement for different cable connectors, compares and optimizes the design ideas, outputs the required binary code through the cascade of multiple encoders, completes the research on the principle of wire number measurement, realizes the signal conversion and display through the single chip microcomputer, and then transmits the tested wire label content to the label printer through the communication agreement, The label marking machine prints out the label from the test results of the wire number measuring tooling, which makes the label printing convenient for use, improves the efficiency of operator’s whole wire bundle labeling, and ensures the accuracy of wire bundle labeling.

Yan Li, Songyin Sui, Qiong Wu, MingHua Zhang, GuangJun Chen, Jun Wang
Correctness Verification of Aerospace Software Program Based on Hoare Logic

Spacecraft software requires more and more credibility. The correctness of program implementation is the basic requirement of software credibility. However, traditional testing methods cannot guarantee the correctness of the program implementation. Therefore, in order to guarantee the consistency between the designing and realization of the software, the method of verifying the spacecraft software by formal method was studied. This paper studied the feasibility of verifying the correctness of the program in theory by formal verification. Firstly, this paper introduced in detail the method of formal verification, and the principles and reasoning rules of theorem proofing based on Hoare logic for part of C language syntax. Then, according to the Hoare logic tripe and its reasoning rules, a module of an aerospace software was verified by artificial reasoning. Finally, it was concluded that the program verification method based on Hoare logic was feasible in the correctness verification of spacecraft software program.

Jian Xu, Hua Yang, Yanliang Tan, Yukui Zhou, Xiaojing Zhang
Research on Image Restoration in Remote Sensing Quick-View System

Quick-view system plays an indispensable row in space exploration and earth observation. Currently, the remote sensing quick-view system of our country only has quick display and store abilities. The motion-blurred remote sensing images would not be restored. This paper centers on image restoration in remote sensing quick-view systems. Through calculating the two-level screw target matching image, the ratio of power spectrum density could be estimated quickly. Then Wiener filtering method is implemented to restore blurred images. To restrain the ringing artifact in quick-view system, different weights are introduced on the edge of the target when implementing Wiener filter. Experiments indicate that the optimized image restoration method can obtain relatively gratifying remote sensing data restorative results and treatment period, which might satisfy the requirement of real time display in remote sensing quick-view system.

Yunsen Wang, Chengzhi Ma, Weiguang Cui, Jiang Liu, Yang Kang
An Image Denoising Algorithm Combining Global Clustering and Low-Rank Theory

In recent years, image denoising algorithms based on non-local image prior information have been extensively researched. At present, most similar image blocks are gathered by neighborhood block matching, that is, a fixed number of similar image blocks are matched in the field of image blocks by a similarity measurement method, basing on Euclidean distance. Although this method is more efficient, there are some problems: when searching for similar image blocks locally, some special structures, such as image edges or corners that cannot match similar image patterns and existing similar structures in the global scope, are ignored.To solve the above problems, a global clustering method is proposed in this paper to match image blocks, which improves the flexibility of the algorithm. At the same time, to solve the problem of the loss of detailed information such as edge texture when the similar image block matrix is constrained by low-rank, a low-rank sparse image denoising model combined with global clustering is proposed to effectively filter out noise while preserving image details as much as possible.

Hongjian Guo, Yaruixi Gao, Nengjian Tai, Xu Yang, Shaoyu Zhang, Xiaobo Hui
Constraint Self-updating Method for Autonomous Task Planning of Earth Observation Satellites Based on In-Orbit Data

The task planning model of earth observation satellite is used for autonomous task planning of earth observation satellite. An accurate model is the premise of efficient optimization of autonomous task planning algorithm. The task planning model generally consists of four parts: optimization objectives, constraints reflecting the real-time state of the satellite system, task input and the data transmission task planning scheme. Task planning constraint is the foundation of the task planning model of earth observation satellite, which reflects the cognition of the satellite capability and current situation. In this paper, for the earth observation satellite with autonomous task planning capability, the modeling of task planning constraints with adjustable parameters is carried out, and the adjustable parameters in the model are estimated in real time by using in-orbit data, so as to effectively estimate the current imaging and data transmission task capability. It solves the problem that the traditional constraint modeling method cannot truly reflect the in-orbit actual state of satellite system capabilities.

Dong Sun, Cong Zhang, Jiaxing Li
A Novel Message Communication Mechanism Based on Partition Name

Space-Time Partitioning software architecture is gradually being used in integrated electronic computers to realize the purpose of running multiple independent application software centrally on the same hardware platform. However, it is impossible to use the traditional data interaction method between the applications under this architecture. In this paper, we propose and implement a message communication method between application software in a Time Partitioning architecture to solve the information interaction problem between partitioned software. Through experiments, it is proved that the method can better meet the loosely coupled software architecture of time-zoning and has good engineering application value.

Peng Fei, Xu Na, Ma Zhi
An Application of Image Denoising Technique Based on Convolutional Neural Network in Star Tracker

Convolution neural network is widely used in the field of computer vision, which can complete the tasks of target detection, image segmentation, semantic generation and so on in complex scenes. As the main optical sensor of the satellite attitude control subsystem, the star tracker is used to calculate the attitude from the star image. In this paper, a local sky area image restoration technique for star tracker based on convolution neural network is proposed and implemented, which can be used to solve the stray light interference, binary star imaging, star point dragging and other problems in the imaging of star tracker, so as to improve the attitude determination accuracy of star tracker. Experimental results show that this method can improve the number of star point recognition and the attitude determination accuracy of the star tracker.

Jingjin Li, Fei Peng, Yizheng Wang
A Time System Based on All Digital Simulation Platform with Precise Clock

The application of full digital simulation platform in the research and development of spacecraft embedded software has the advantages of fast construction, low cost, rich debugging and testing methods and supporting fault injection. However, due to the difference of processing ability, real-time difference and simulation speed, the traditional digital simulation platform based on host time system can’t meet the requirements of real-time certainty of embedded software task, which will cause the software operation confusion. Therefore, this paper proposes a time precise time system based on instruction cycle. Based on the execution cycle and basic block of SPARC V8 instruction set, the algorithm and implementation of time system are given considering the translation execution efficiency, timer interrupt and execution abnormality. Compared with the results of the simulation system running on the full digital simulation platform and the hardware platform, the time system proposed in this paper has the high efficiency of time simulation, and the time error rate is low, which can meet the needs of the actual model application.

Ruijun Li, Tao Zhang, Hongjing Cheng, Yanfang Fan, Jiali Zheng
Medium Power Class-E Amplifier Design

Switch-mode amplifiers like class-E and class-F have high efficiency and more application in RF system. The high efficiency amplifiers are normally used in the last stage of the power amplifier, but along with more and more critical DC power requirement, driving stage (medium output power) of the power amplifier also requires high efficiency. Based on the class-E amplifier operation and design principle, a 2 GHz medium output power class-E amplifier was designed and manufactured. The test results show agreement with simulation and the amplifier has peak PAE of 78%.

Lin Pan, Xiaoran Li, Gang Dong
Ka Band Satellite Communication Adaptive Coding and Modulation Technology Based on Artificial Intelligence LDPC Code

Due to the rapid development of broadband satellite communication technology, the previous C band and Ku band can no longer meet the full range of needs in the communication field. Ka frequency band has outstanding performance with large bandwidth and good anti-interference ability, which can make Ka frequency band a commonly used frequency band in the field of satellite communication. However, rain attenuation is a key factor affecting Ka band satellite communication. In order to make the signal stable and accurate transmission in the satellite communication process and overcome the adverse effects caused by rain attenuation, it can be achieved through adaptive technology. In the context of artificial intelligence, this paper studies the adaptive coding and modulation technology based on LDPC codes in Ka band satellite communications. This article first made relevant research on Ka band satellite communication, and then designed an adaptive coding and modulation transmission mode. Finally, the performance of the maximum likelihood algorithm and the second-order-fourth-moment algorithm are analyzed through simulation experiments, and the normalized estimation errors and estimated variances of the two algorithms are compared, and the corresponding results are obtained according to the experimental results.

Guozhi Rong, Yikai Wang, Gang Dong, Yifeng He
Encoding and Decoding Algorithm of LDPC for Low Orbit Satellite Communication Based on Artificial Intelligence

As the most promising satellite mobile communication system at present, the low-orbit satellite communication system has the advantages of small delay, low path loss, and wide coverage. Low-density parity-check (LDPC) code, as an emerging channel coding, can provide high coding gain while maintaining high coding efficiency, which is very suitable for low-orbit satellite communications. The purpose of this paper is to study the encoding and decoding algorithms of LDPCs for low-orbit satellite communications based on artificial intelligence. This article introduces the basic coding method of LDPC and the original model LDPC in detail, and explains its advantages over LDPCs. Based on LDPCs, the coding algorithm is improved. This paper proposes an early stopping iterative algorithm and further simplifies the LLR decoding algorithm. In the low-orbit satellite communication environment, the LDPC encoding and decoding algorithm proposed in this paper is simulated, and the influence of interference factors in the actual environment on the performance of the LDPC is tested. The simulation results show that under the Rice channel condition, when the Rice coefficient is greater than 8, when the bit error rate, the bit error rate coding gain of LDPC decreases by about 2.8 dB, which proves that the algorithm proposed in this paper can be used in the low-orbit satellite communication environment.

Yifeng He, Yijian Zheng, Jingyi Zhang, Wenze Qi, Dayang Zhao
Capacity Analysis of Relay Channel Based on Computer Technology and LDPC Coding Cooperation and Satellite Communication

Information theory, channel coding and cooperative communicate`on are the three theoretical pillars of modern digital communication systems. Information theory is based on revealing the essential characteristics and laws of information, and its basic task is to provide a theoretical basis for the design of effective and reliable communication systems. Channel coding is developed and gradually matured under the guidance of Shannon's information theory. It provides an effective means for reliable transmission of information in the channel. Among them, the LDPC code has become one of the most popular research directions in the current channel coding field due to its excellent performance close to the Shannon limit and huge potential application value. This article first quickly popularizes the concept of digital communication systems, analyzes the LDPC code technology and cooperative MIMO technology in detail, then conducts an in-depth study on the capacity of the relay channel, and finally combines the LDPC code with the cooperative communication, and proposes the LDPC coded cooperative communication Program. Finally, the LDPC code is applied to satellite communication, and the performance of the satellite communication system is improved through different methods.

Rui Zhang, Shuanglan Mao, Nan Ye, Bo Zhang, Liang Peng, Zhihang Mao
Design and Verification of HY-2A Satellite Precision Orbit Determination Technology

This paper discusses the HY-2A satellite precision orbit determination system. The satellite-to-ground system configuration, target allocation, and working modes are elaborated separately. The specific implementation methods of precise orbit determination technology under different working modes are analyzed. Finally, the actual results achieved in the on-orbit mode are given, and the application prospects of precision orbit determination technology are discussed.

Jie Liang, Yi Fang, Shilin Dong
Design of High-Reliability LDPC Code Algorithm for Relay Satellite Communication System

This paper introduces a hardware platform and software algorithm for the design of relay satellite LDPC encoding and decoding, And use computer simulation method to simulate the performance of LDPC (8176, 7154) code normalization minimization decoding algorithm, determine the correction factor of the decoder through analysis and simulation verification. Compared with the uncoded system, the system can obtain a coding gain of about 7.0 dB, which is consistent with the index results given in the CCSDS standard.

Xiaoran Li, Pan Lin, Yifeng He
Research on Real-Time Monitoring Method of Spot Beam Antenna Direction on Satellites

Since the traditional test method of satellite takes too long to judge the pointing direction of the spot beam data transmission antenna, which affects the satellite test process, a real-time monitoring method of satellite spot beam antenna based on real-time input and real-time output (RIRO) is proposed. The real-time data from the satellite test system is used, and the status of the real-time direction of the antenna is checked through the three-dimensional display. The new method is applied to GF-3 satellite and a real-time monitoring system is designed to monitor the antenna direction in a visual way, and to interpret the pointing accuracy in real-time, which greatly improves the test efficiency and real-time interpretation ability of the satellite. The method in this paper can be applied to satellites including spot beam antenna.

Zhi Yuan, Lina Wu, Yinghua Li
A Novel SOC Estimation Approach for the Lithium-Ion Battery Pack Using in the Deep Space Landers

Lithium-ion battery pack has been widely implemented in the spacecraft as the crucial energy supplying unit to accomplish the exploration task of deep space landers. State of charge is a key index to indicate the present available capacity of the battery pack, which barely can be measured by precision instruments due to the nonlinearity and time variation of the battery. In this paper, a model-based method is studied to estimate the SOC of the battery pack. It is demonstrated to be effective by actual operating cycles with the accuracy under 3% estimation errors.

Hao Mu, Zhigang Liu, Wang Jing, Dong Yang
Analysis of Characteristics and Requirements of Large LEO Constellations Management

With the rapid development of micro-satellite technology, the significant reduction in the cost of mass satellite production, and the rapid growth of application demand in the fields of global communication, navigation and remote sensing, large LEO constellations with the advantages of global seamless coverage, low transmission delay, high transmission rate, large information capacity, and individual satellite failures do not affect overall performance have become a major hot spot at present and for quite a long time in the future. As a basic support for constellation construction and effectiveness, how to manage a large constellation stably and efficiently has become a practical problem that needs to be solved urgently. Briefly introduced the development of large constellations domestic and abroad, sorted out and summarized characteristics of management and challenges faced in long-term operation compared to single satellite or small constellations, analyzed and proposed management requirements such as automation and intelligence, health management, configuration maintenance, and autonomous operation. Research results can be used as a reference for the development and construction of large constellation management systems.

Da Tang, Guoting Zhang, Tang Li, Wanhong Hao
Research on Space Project Systems Technological Risk Assessment Based on Multi-level Fuzzy Comprehensive Evaluation

It is of vital importance to develop systemic technological risk assessment for a space project. Combined technological risk character with basic architecture, quantitative evaluation index system, risk level criteria, and quantitative method is established for space project technological risk, and then, quantitative assessment model for systemic technological risk is proposed based on multi-level fuzzy comprehensive evaluation algorithm, which facilitates analysis, assessment and management of space project technological risk.

Lily Wang
Research on the System Construction of Spaceborne Adaptive Software and Its Key Technologies

With the deployment of large-scale satellite constellation, the complexity and scale of spaceborne control software have shown remarkable increases. The operating environment has also shown dynamism and variability. Constantly changing software requirements require higher levels of flexibility and scalability of the software system in an open environment. However, it is difficult to dynamically change the variability model for most spaceborne control software, leading to difficulties for the system to adapt to required environments for desired changes during an operation. The objective of this study was to develop a construction technology for the adaptive system of the spaceborne control software. Firstly, this study designed an overall framework of the target software system with its key functions. We established a unified description model of variability such as working mode and uplink injection to support software dynamic adjustment. We then unified the API using the design message middleware for hardware assessments. Furthermore, adaptive software parameter adjustment mechanism was discussed in this study. We extracted behavior parameters to form independent external configuration files and implant the connector with data flow constraints in functional units. Finally, an adaptive strategy was designed to exhibit an adaptive adjustment effect on the behavior and structure made by the adaptive system under the change of demand for certain satellite control software.

Xiaofeng Li, Qian Wu, Xiaolong Yang
Air-Space-Ground Integrated Information Network: Technology, Development and Prospect

As a kind of complex heterogeneous network with broad application prospects, the air-space-ground integrated information network is a focus of research and construction investment in recent years no matter from the perspective of application or technology. In this paper, based on the research and construction status of the air-space-ground integrated information network, the definition, composition, key technology and development status of the air-space-ground integrated information network are analyzed, and the future research direction is prospected. This paper can be used as a reference for research in related fields.

Gang Dong, Lin Pan, Yi Zhang, Yanan Liu
High-Resolution Satellite Radiation Big Data All-Statistical Intelligent Screening Technology

At present, domestic high-resolution satellites for civil use the radiation big statistical data technology to carry out relative radiation calibration of the optical load. Aiming at the screening of high-end, low-end and cloud cover in big data statistics, this paper proposes a method based on probability and single value negation. The strategy of data screening ensures that the statistical data better represents the statistical characteristics of the optical load radiation, and achieves the improvement of the relative radiation calibration quality.

Huimin Zhong, Xi Wang, Wenyong Yu, Xiaoxiang Long, Xiaoyan Wang
EMC System Control on XX-2 Satellite

Surrounding with the key EMC work emphasis on satellite system, this article mainly introduced the main EMC work content and flow of EMC system control on XXx-2 satellite. The main work includes EMC work planning, establishing an EMC working group, stand-alone equipment EMC control, satellite system EMC analysis and prediction, Radiated Model (RM) satellite test, EMC system test verification and so on. In this paper the methods and steps of each part of the work are briefly introduced, as well as the specific effects on the control of the electromagnetic compatibility of the satellite. Some electro-magnetic interference (EMI) problems discovered in the process of development and their solutions are introduced.It is proved that the EMC system control work has been included in the whole RD period for xx-2 satellite, which has broadened the idea of satellite EMC work, ensures this satellite task is completed successfully. It also provides references and development experiences for other new microwave payload aircraft’s emc work in the future.

Liping Zhou, Hua Zhang, Lili Cheng, Yajie Zhang
Transmission and Control of the Image Data from Remote Sensing Satellite to the Ground Station

The image data of remote sensing satellite are widely applied all over China and play an important role in economic and social development and ecological civilization construction. Based on the insufficient established ground station of China, the big data of remote sensing satellites should be processed suitably in orbit, compression and distribution are necessary, and be transmitted to the ground rightly and efficiently. The data quantity and the mode of the data transmission are analyzed. The composition and the mechanism of the data transmission system are introduced. The pointing control of antenna determines the rightness and efficiency of the transmission. Requirement of antenna control is analyzed and the way to weaken the disturbance from antenna is proposed. Meanwhile, the pointing control logic of antenna is described and the angle curve of antenna in orbit is given and analyzed.

Xiao-ming Wang, Xiao-dong Wu, Bo Han
Micro-vibration Attenuation Design, Analysis and Verification for Agile Remote Sensing Satellite

Micro-vibration sources such as control moment gyroscope (CMG) on agile remote sensing satellite can cause micro-vibration, affecting the performance of sensitive payloads. Pointing at this problem, and based on the micro-vibration attenuation requirement of GFDM-1 satellite, micro-vibration attenuation design, analysis and verification are performed. The methods shown in this article can provide reference for the micro-vibration attenuation for the ongoing agile remote sensing satellite of China.

Xingsu Gao, Guangyuan Wang, Yongsheng Wu, Yue Wang, Yu Zhao
Practice and Research on FMEA of Telecommunication Satellite System

In recent years, telecommunication satellites have developed in the direction of large load-bearing ratio, high integration, and large load capacity. At the same time, with the substantial increase in propellants, higher requirements have been put forward for the long life and high reliability of satellites. System-level reliability work is the focus of satellite model product assurance work, and failure mode and impact analysis (FMEA) is the core of satellite system-level reliability work. Through a series of effective practical methods, the author carried out system FMEA analysis for a commercial communication satellite. and then obtained some existing problems and suggestions through the problems encountered in the practice. In this process, we have gained valuable experience and put forward some suggestions at the end of article.

Hao Zhang, Zong Ren Wang, Xue Wang Wang, Feng Chun Lin
Characterization of On-Orbit Temperature and Pressure Performance for Hydrazine Propulsion System

To characterize on-orbit temperature and pressure variation of the hydrazine propulsion system, three types of temperature telemetries including the catalyst bed temperature/tank temperature/pipe temperature, and two types of pressure telemetries including tank pressure/pipe pressure were investigated. Results indicate that the catalyst bed temperature of the 5N hydrazine thrusters decline in the first few minutes before going up during the preheating process, and the turning point is related to the separation of the rocket fairing. The decline of the catalyst bed temperature is owing to the heat convection between the thruster and air in the fairing. The pressure fluctuation during thruster firing is in the range of ±0.25 MPa, but pipe pressure rises up apparently with a speed of 0.5 MPa/°C even the pipe temperature rises slightly in the locked pipe. Accordingly, monitoring measures of the temperature and pressure variation should be taken during on-orbit management of the satellite.

Yong Gao, Yuan Wang, Ning Yao, Jialong Ji, Jiaai Yang
On-Orbit Calibration of a 7-DOF Space Arm

For space manipulator, high control precision is required to realize precise operation of the target object, which directly depends on the measurement of the structure parameters. However, because of the concussion during launch and the different gravity on-orbit, the structure parameters of the space manipulator can not keep the same as on the ground. Therefore, it is important to recalibrate the space manipulator on-orbit, which becomes difficult because high precious measurement devices such as theodolite can not be used. In this paper, an on-orbit calibration method is proposed. First, the error model of the structure parameters is established. Then, the orbit calibration data is measured by the hand-eye cameras. Finally, the structure parameters can be achieved by solving the error model with least square method. The proposed method can solve the errors of the hand-eye cameras related to the hand end, and it can improve the precision of calibration.

Rui Liu, Shuai Xiao, Chi Zhang
Correlation Analysis and Forecast of Electric Vehicle Charging Stations Considering Seasonal Factors

With the widespread use of electric vehicles, charging stations have become the critical infrastructure to ensure the operation of EV. Reasonable optimization of EV charging stations during the planning stage could contribute to save customers’ time, reduce equipment loss and reduce grid load. The analysis results show that the customer arrival interval of the charging station approximately obeys the negative exponential distribution, and the charging service time approximately obeys the gamma distribution, and the variation curve of the distribution parameters with the ambient temperature is given. A queuing simulation model for charging stations was established. The Monte Carlo method was used to analyze the queuing length and time of charging stations in different seasons, which provided a reference for the construction and optimization of EV charging piles in the future.

Qianxu Yu, Jing Ren, Shanshan Wei
Inversion Modeling Study of Dissolved Oxygen in Wangyu River Based on BP Neural Network

To explore the application potential of BP neural network model in the retrieval of river water quality parameters from high-resolution multi-spectral satellite images. This paper takes Wangyu River as the research object, based on the GF-1 and GF-2 images and measured dissolved oxygen data. Two models of BP neural network and linear regression were used to simulate the dissolved oxygen content of rivers and the fitting effects of the two models were compared The results showed that: for dissolved oxygen, the average measured relative error of BP neural network and linear regression models is 26.2% and 61.03%. The inversion effect of BP neural network is better, and the linear regression model has larger prediction errors for individual monitoring points. Combining high-resolution remote sensing stars and BP neural network model can effectively monitor the water quality of rivers in the city.

Min Jin, Xinhua Qian, Lei Wu, Yuhu Li, Bangguo Hu, Wangmeng Chen
Research on Satellite Energy Autonomous Security Strategy

With the rapid development of aerospace technology, satellites show the characteristics of high net worth, long life, and complex operating conditions, and the importance of safety in orbiting satellites has become increasingly prominent. Based on the current mission requirements of domestic satellites, this article outlines the architecture, safety, and autonomous management of the energy system. Combined with the requirements of long life and high reliability, this paper proposes a satellite energy autonomous security management architecture design plan, focusing on power supply security The discussion and evaluation of power distribution safety, energy margin safety, etc. can provide reference and reference for the realization of the subsequent satellite autonomous management functions.

Xiaofei Li, Ming Qiao, Bingxin Zhao
Research and Development of High Speed and Large Bandwidth MIMO Data Transmission Test System

Based on the modular architecture to build the transceiver communication hardware platform, this paper develops a high-speed and large bandwidth MIMO data transmission test system. This paper uses two pairs of L/S band transceiver antennas, reasonably designs the frame structure, and realizes the 5 Gbps high-speed information wireless transmission function of the whole system through MIMO space-time coding, synchronous equalization and other physical layer data processing technologies. This paper fully considers the characteristics of hardware equipment and the characteristics of actual wireless channel, studies the data frame structure and baseband algorithm design, and puts forward the research idea of hardware verification high-speed data transmission under the background of large bandwidth. The built hardware system can verify the wireless channel characteristics in a variety of application scenarios; The algorithm is unitized and can verify a variety of communication models. Through the actual test of hardware equipment, the rationality and reliability of the system design are verified.

Chai Yuan, Shuai Han, Zhiqiang Li

Multimedia and Communication

Frontmatter
Continuous Evolution for Efficient Neural Architecture Search Based on Improved NSGA-III Algorithm

Improved Continuous Evolution for Efficient Neural Architecture Search method (I-CARS) is proposed to solve multi-objective optimization problems (MOPs). Seeking to improve the convergence and search accuracy, two modifications of the non-dominated sorting genetic algorithm based on reference-point strategy (NSGA-III) were made to replace pNSGA-III in CARS, including the penalty-based boundary intersection distance (PBI distance) and the selection-and-elimination operators. First, the perpendicular distance from solutions to reference lines was replaced by the PBI distance in the offspring selection stage phase, which can add the convergence information. Second, individuals in the population were selected or eliminated by niche count and PBI distance. Better performing individuals were selected to become the next generation, while poorly performing individuals were eliminated. The convergence and diversity of the population can be balanced by adding the selection-and-elimination operator. Experiments were conducted on Vega pipeline, and I-CARS achieves 3.41% test error on CIFAR10, the results indicated that the accuracy and convergence of I-CARS are enhanced compared to CARS.

Ziyan Wang, Feng Qi, Liming Zou
Research on Individual Identification of Radar Emitter Based on Multi-domain Features

Individual identification of radar emitter is an essential part in electronic warfare. With the increasing complexity of electromagnetic environment and the rapid development of electronic equipment technology, the feature extraction method based on traditional pulse descriptor has been unable to perfectly meet the requirements of current emitter identification. Based on real radar emitter signal and combined with signal transform feature extraction and automatic feature extraction of autoencoder, this paper proposes a method of extracting intra-pulse features of radar emitter based on bispectrum, wavelet packet decomposition and time-domain statistical features. Decision tree, KNN and SVM algorithms are used to classify the radar emitter based on the extracted signal features, and classification accuracy of decision tree is close to 95%, which means the individual identification of radar emitter is realized perfectly.

Bo Li, Songlin Sun, ZhiKai Su, Chenwei Wang
Fast HEVC Intra CTU Partition Algorithm Based on Lightweight CNN

Recently, several methods have been proposed to reduce the complexity of high efficiency video coding (HEVC) by using neural networks to replace the original rate-distortion optimization (RDO) search for obtaining the optimal coding tree unit (CTU) partition. However, these approaches are difficult to deploy in scenarios with limited storage space due to the complex network structure. In this paper, we propose a lightweight convolutional neural network (L-CNN) for fast CTU partitioning with only a KB-sized model, which makes it applicable in storage space-constrained scenarios. First, we establish a dataset that contains spatial information and image texture for each coding unit (CU). Then, we combine CU spatial information with image data to obtain the CTU partition to further improve the performance of the network. Finally, experimental results show that our method reduces the model size by 13.7× and 5.2× compared with two CNN-based approaches with 60.46% intra-mode encoding time reduction and 2.78% Bjøntegaard delta bit-rate degradation, outperforming the existing state-of-the-art approaches.

Rifa Zhao, Hai Huang, Ronghui Zhang, Xiaojun Jing
A New Pulmonary Nodule Detection Based on Multiscale Convolutional Neural Network with Channel and Attention Mechanism

In this stage of pulmonary nodules image processing, the pulmonary nodule detection and recognition based on 3D CNN has achieved great success in medical image processing. The tradition pulmonary nodule detection model has a low detection accuracy, in order to more accurately detect the pathological pulmonary nodules, this paper proposes a novel 3D CNN model based on multiscale architecture with channel and spatial attention. That is to say, multi-scale feature extraction module and attention mechanism are added to some residual blocks in 3D Fast-R-CNN, which can improve the performance of the whole network and better learn the features of candidate nodes; Secondly, in the part of regional proposal network, U-Net network and Fast RCNN are combined. The method on LUNA16 dataset shows that the average free-response receiver operating characteristic (FROC) score is 0.848 which is better than other methods. The proposed method can well reduce false positives and improve detection accuracy, providing a reference for clinical medicine.

Yingying Zhao, Jiaxin Wang, Xiaomin Wang, Honglin Wan
A Novel Attention Mechanism for Scene Text Detection

In recent years, segmentation-based methods have been developed rapidly in scene text detection, but there is still a challenge in detecting text with irregular shapes and large size variations. To stress the issue, this paper proposes a novel multi-scale spatial attention fusion network based on backbone (MSAFNet). The network includes a backbone and multi-scale spatial attention fusion (MSAF). The MSAF can enhance the feature extraction of small scale text and increase the feature extraction of large scale text. Experimental results on ICDAR2015 depict that the proposed method is more effective than the most advanced text detection methods.

Jiaxin Wang, Yingying Zhao, Xiaomin Wang, Honglin Wan
Epileptic Seizure Detection Based on Multi-synchrosqueezing Transform and Multi-label Classification

Machine learning is a significant tool in the study of epileptic automatic detection system. This paper presents a novel method based on multi-synchrosqueezing transform (MSST) and multi-label classification (MLKNN). This proposed algorithm is developed on the basis of MATLAB software. Firstly, the processed EEG data is transformed by MSST. On the basis of synchronous compression transform, iterative redistribution is adopted to replace the original EEG data to complete pattern recognition. Then input the relative mean of the feature index and amplitude into the MLKNN classifier. Finally, the automatic detection of epileptic wave is realized through post-processing. The method was evaluated on the EEG data in Boston Children’s Hospital, and the results were satisfactory. The accuracy for both episodic and episodic EEG on the Boston database was 95.77%, with a mean sensitivity of 86.31% and specificity of 87.33%. Compared with the traditional automatic detection model based on machine learning, the performance of epilepsy detection is improved.

Yuying Yang, Jiazheng Zhou, Jianping Liu, Qi Yuan
Mobility-Aware Task Offloading Scheme Based on Optimal System Benefit in Multi-access Edge Computing

Multi-access edge computing (MEC) is an effective technique to provide computing resources in close to users. However, the offloading scheme for computing tasks in the high-speed movement scenario is usually complicated and intractable. Aiming at the problem of selecting the target MEC server for task offloading in the high-speed movement scenario, a MEC server selection scheme based on the optimal system benefit is designed by defining the system benefit function (O-CCE) in this paper. Specifically, considering the influence of connection time between mobile user and MEC server, a service processing cost model (TRCMI) of the MEC system is constructed. Comprehensive consideration of communication resource, calculation resource, energy consumption, connect time, and amount of task, the best MEC server is selected to provide services for users, and the optimal task offload is also given. The simulation results show that this solution can improve the system efficiency and achieve the overall optimum with limited resources.

Chuanfen Feng, Jiande Sun
Design of Electronic Intelligent Code Lock

The research is mainly designed for the current people’s demand for electronic code locks. Its main functions include code unlocking, IC card unlocking, remote mobile phone monitoring, random password unlocking etc. The main control chip of the intelligent electronic code lock adopts STM32 single-chip microcomputer. It uses the H35B-UC LCD touch screen to display the main interface of the system. It uses the RC522 radio frequency card module to read and write the IC and realizes the function of opening the door of the IC card. Moreover, it is connected to the Internet through the ESP8266 WIFI module, which can realize the remote control of the smart electronic code lock. The button module does not use external independent buttons, but through the H35B-UC LCD touch screen to achieve digital password input. It is suitable for ordinary family door locks and hotel door locks etc.

Qian Bao, Zixuan Wang, Xinpeng Wang, Zhixin Guo, Deyan Wang, Jia Guo, Chunxing Wang
Integrating Ensemble Learning and Information Entropy for Diabetes Diagnosis

GDM (gestational diabetes mellitus) is a common disease during pregnancy, it will affect the amniotic fluid secretion and immune system, leading to premature delivery and even early stripping of the fetus. So, whether GDM can be sent out timely and accurately is very important.In recent years, due to the great breakthrough of machine learning algorithms in the medical industry, many scholars have applied it to GDM diagnosis. However, due to the high dimension of GDM data features and the small amount of data, Most algorithms have poor generalization and stability. Catboost model is the best machine learning model in GDM diagnosis. Compared with other machine learning models, it can make assumptions about the differences between the actual samples and the real samples to help predict. So it is widely used in GDM diagnosis scenarios and achieved good results. However, there are two main problems in the existing research: 1. Most of the research in the post-processing steps of the model threshold selection is not rigorous enough, which can not highly match the information entropy of the original data. 2. The performance of a single catboost model will gradually decrease when the number of data increases. In order to solve the problems above, This paper takes catboost as the core to build an integrated learning algorithm, and innovatively takes the information entropy of the original data into account when determining the prediction threshold of the model. Our method achieves better results than traditional methods on the open source GDM dataset.

Cheng Li, Yiyang Xiong, Xuezhi Zhang, Ruitong Liu, Xiaojun Jing
The Optimal UDP Data Length in Ethernet

The network has become an integral part of people’s daily life. The reason why communication network can realize information exchange and transmission is largely based on the collaborative work of network protocols and routing algorithms, and some hardware facilities. The network protocols and routing algorithms are important elements in designing communication network systems. TCP (Transport Control Protocol) and UDP (User Datagram Protocol) are two transport protocols of the transport layer of the OSI (Open System Interconnection). UDP protocol has the characteristics of high efficiency and fast transmission, while TCP protocol has a slow transmission rate and takes up overhead. Therefore, UDP protocol is used in the communication systems with high requirements for real-time transmission, such as multimedia communication system. In the process of transmission, in order to reduce the burden of CPU on protocol processing, when the length of a UDP segment exceeds the MTU (Maximum Transmission Unit), the IP packet will be fragmented, and the transmission efficiency will be reduced. However, if the length of a UDP segment is too short, the number of UDP segments needed to transmit the same file will increase, which will also limit the transmission efficiency. This paper builds an AVS2 video streaming on-demand system in the Ethernet environment. By changing the length of the UDP data, the video transmission time and bit loss rate will be obtained and used to choose the optimal UDP data length to make the transmission efficiency the highest.

Chunhao Li, Lingfeng Fang, Songlin Sun, Tingting An, Chenwei Wang
An Attention-GRU Based Gas Price Prediction Model for Ethereum Transactions

Each operation on the Ethereum network requires a corresponding gas cost, and the gas price is customized by the user who uses it. Under normal circumstances, the shorter the user’s requirement for the transaction completion time, the higher the gas price should be set. However, users cannot set a reasonable gas price every time, which will cause operation failure or gas waste. Therefore, predicting an affordable gas price is of great significance. We propose a gas price prediction method based on deep learning to find the lowest transaction gas price in the next block and help users save transaction costs. Based on previous research, we obtained influence factors of block gas upper limit, transaction gas upper limit and mining difficulty that cause gas price fluctuations, combined with the attention mechanism to obtain the weight of the influence factor, we use the gated recurrent unit (GRU) neural network to determine the gas price prediction. Extensive experiments based on 2000 actual blocks show that the proposed gas price prediction method has high accuracy and reliable performance, which saves users a large amount of transaction costs; because the appropriate gas price can be predicted, the success rate of users’ transactions increases by 22.9%.

Yunxia Feng, Yuan Sun, Jing Qu
A Research on an Unknown Radar Signal Sorting with DBSCAN-BP Algorithm

For the unknown radar signal, this paper proposes a radar signal sorting algorithm DBSCAN-BP algorithm which combines unsupervised learning with supervised learning. DBSCAN algorithm is used to pre-sort and de-noise the unknown radar signal pulse data, and then BP neural network is used to finally sort the pre-sorting results of DBSCAN algorithm to realize the radar signal classification. In addition, this paper also proposes a method of radar data set division based on frame number, which can guide the simulation of real scene and make the radar sorting process more scientific.

Zhikai Su, Songlin Sun, Yuhao Liu, Chenwei Wang
An Industrial Internet Delay Optimization Scheme Based on Time Sensitive Network

In order to improve the real-time performance of Industrial Internet, this paper proposes a network communication delay optimization scheme that based on the IEEE 802.1Qbv protocol of time sensitive network (TSN). The scheme distinguishes the network data types with different time delay requirements, and proposes a TAS-EWRR scheduling algorithm in the data link layer of OSI model. The algorithm combines with time aware shaper (TAS), stream reservation protocol (SRP), frame preemption mechanism and enhanced weighted round robin (EWRR) algorithm. At the same time, a shortest path algorithm based on delay is combined at the network layer to realize the optimization of the delay of the entire network.

Lei Wu, Ju Liu, Tianfeng Xu, Xueping Shao
The Key Technology of Ultra-high-Speed Laser Communication Principle Verification System

The satellite platform simulator, OOK system optical communication link modulation/demodulation and data storage, and support for the user terminal simulator are needed at each stage of the inter-satellite laser communication terminal in the development process. In order to simulate the entire inter-satellite laser communication link, we designed and verified a complete set of data transmission procedures to meet the test requirements of different functions of inter-satellite laser communication terminals at various stages. This system realized the transmission and collection of real-time user data. The real-time storage data rate could reach up to 2.4 Gbps, and the sensitivity of the received optical power of the system was verified by the bit error rate test, which could reach −32 dBm. In order to adapt to changes in different space environments, we had set a mode where the user data transmission rate could be switched, and the switching response time was less than 100 us.

Shidong Lv, Yilong Bai, Xinghua Wang, Jinghua Wang
Inappropriate Visual Content Detection Based on the Joint Training Strategy

In the information age, massive Internet data brings convenience to us. But there is some inappropriate visual content (pornography, violence, politics, terrorism, etc.), among which the dissemination of pornographic content has an adverse influence, especially for children and minors. Therefore, we present an inappropriate visual content detection method based on the joint training strategy in an end-to-end manner, which realizes the identification and location of inappropriate visual content while retaining the base class (80 categories in the COCO dataset) detection. To solve the difficulty of sample labeling, in this paper we propose a combined training strategy of detection and classification. And the Focal loss is used to improve the sample imbalance in the network sharing training. The algorithm can achieve multi-label output and has good recognition accuracy. Finally, a more challenging dataset INVC of inappropriate visual content is proposed, which includes three types of sample data in complex backgrounds at different scales, such as indoor, beach, street, etc.

Xuejing Wang, Ju Liu, Xiaoxi Liu, Yafeng Li, Luyue Yu
Spectral Optimization for Multi-carrier NOMA-Enabled Industrial Internet of Things

Non-orthogonal multiple access (NOMA) guarantees good spectral efficiency and better enhance system capacity. However, the deployment of NOMA is not easy due to complexity of resource management. This paper provides a new resource optimization architecture to improve the spectrum efficiency (SE) of the multi-carrier NOMA enabled Industrial Internet of Things (IIoT) network. Specifically, we jointly optimize carrier-user assignment and power control to maximize the SE of the system while satisfying the constraints on users quality of services (QoS) requirements, transmit power budget and successive interference cancellation. Since the optimization problem is a mixed integer non-convex programming problem, it is very difficult to solve. We decouple it into two sub-problems to solve. Sub-problem one: subcarrier-user assignment problem; Sub-problem two: power allocation to users on each subcarrier. Particularly, two subcarrier-user assignment algorithms (SUAA1 and SUAA2) are proposed to solve sub-problem one. In which SUAA1 is based on the traditional NOMA user pairing scheme, SUAA2 is based on strong users first. The sub-problem two is solved based on the Monotonic optimization method. The simulation results show that our resource management scheme has higher SE comparing with OMA scheme. Specifically, SUAA2 can improve the SE of the system while increasing the rate of far users comparing with SUAA1 and the random matching scheme under our power allocation algorithm.

Tianfeng Xu, Ju Liu, Zhichao Gao, Lei Wu, Shanshan Yu, Xueping Shao
Local Edge Structure Guided Depth Up-Sampling

The current depth up-sampling cannot effectively using the pixel correlation and the edge structure between depth map and color map. In this paper, we proposes a new depth up-sampling method based on the local edge structure. Firstly, the non-robust pixel refinement in the low resolution (LR) depth image is obtained from the LR sobel-based edge map. Then, the structural consistency judgment within the depth map and the color map and the pixels classification is implemented with the guidance of the gradient matrix of the high resolution (HR) color image. Finally, according to the influence between pixels and spatial location constraints, the depth map is refined by the effective depth. Extensive experiments demonstrate that the proposed method outperforms conventional interpolation algorithms and some other edge-based depth up-sampling methods.

Chunxing Wang, Zeying Xue, Lanjing Zu, Wenbo Wan, Zixuan Wang, Zhixin Guo, Yannan Ren
Deep Learning and XGBoost Based Prediction Algorithm for Esophageal Varices

The deep convolutional neural network has been extensively applied for clinical computer-aided diagnosis. In this study, we combine deep learning feature extraction and eXtreme gradient boosting (XGBoost) classifier for predicting the risk of esophageal varices (EV). First, the quantitative deep learning features and radiomics features of the regions of interest which includes spleen, liver and esophageal are extracted and concatenated. Then, XGBoost and the Least Absolute Shrinkage and Selection Operator (LASSO) are applied for the optimal predictive features selection and prediction of EV risk. XGBoost is used to assess the significance of the extracted features and LASSO is used to select the distinctive features. Finally, random forest, XGBoost and support vector machine classification methods are applied for predicting the low-risk and high-risk of esophageal varices. We collected computed tomography images of cirrhotic patients in two hospitals as the independent training and validation sets. Experimental results show that the features of esophageal are more distinctive than that of other organs. Moreover, the combination of deep learning and radiomics features based on XGBoost algorithm has outperforming classification performance in predicting the severity of EV disease compared to existing approaches.

Xinyi Chen, Jiande Sun, Zhishun Wang, Yanling Fan, Jianping Qiao
A Data-Driven Model for Bearing Remaining Useful Life Prediction with Multi-step Long Short-Term Memory Network

Accurate prediction of the remaining useful life of rolling bearing is of great significance for making reasonable maintenance strategy and reducing maintenance cost. In order to extract the features with time sequence information better for prediction, we propose a data-driven method using multi-step long short-term memory (MS-LSTM) network for predicting the remaining useful life (RUL) of bearings. Firstly, stacked denoised autoencoder (SDAE) and self-organizing maps (SOM) are combined to construct one-dimensional health index (HI) curve according to the original vibration signal, and then the HI curve is input into the MS-LSTM network to predict the long term future trend. Finally, the remaining useful life is calculated according to the failure threshold. Compared with the existing advanced methods, it achieves an absolute improvement of RMSE by 13.63% in the whole remaining life prediction and 7.43% in the last 840 steps.

Zhen Zhao, Zehou Du, Kaixuan Yang, Haoqin Sun, Jialong Wei, Yang Liu
State Estimation of Networked Control Systems with Packet Dropping and Unknown Input

This brief considers the problem of state estimation for networked control systems with malicious packets drop between controllers and actuators. We firstly take advantage of the statistical characteristics of packets drop to estimate the system state. However, when it is difficulty to obtain the statistical characteristics of packet drops, we transform the uncertain malicious packet drops estimation problem into an unknown input estimation problem, then we design a recursive estimator, in which the unknown input estimator is obtained from the innovation analysis and the state estimate is derived by a procedure which similar to the Kalman filter. Finally, we utilize simulation examples to show the effectiveness of the estimators.

Lei Tan, Xinmin Song, Guoming Liu, Zhe Liu
Self-attention Mechanism Fusion for Recommendation in Heterogeneous Information Network

In recent years, lots of recommendation methods based on heterogeneous information network has been proposed. However, traditional heterogeneous information network mainly relies on the similarity of meta-path to recommend, which cannot fully mine the user’s potential preference features. We propose a heterogeneous information network model using self-attention mechanism for recommendation (HINS). In our method, local neighbor information is modeled for users and items respectively via collaborative attention mechanism. Moreover, by optimizing the relationship representation between learning nodes of multi label classification problem based on meta-path, self-attention mechanism is added to the relationship representation to make the model personalized. Finally, the two parts are unified into one model for top-N recommendation. We do evaluation experiments on four different datasets, and our model achieves the best results.

Zhenyu Zang, Xiaohui Yang, Zhiquan Feng
DAMNet: Hyperspectral Image Classification Method Based on Dual Attention Mechanism and Mix Convolutional Neural Network

Deep learning approaches have been broadly used in the study of hyperspectral images (HSI) with promising results in recent years. However, improved network complexity and features extraction still have scope for improvement. As a result, we propose a hybrid convolutional neural network (DAMNet) based on a 3D-dual attention mechanism (spectral and spatial attention mechanism) to fully capture spectral and spatial features and improve classifying precision. This is a spectral-spatial 3D-CNN that is being followed. The 3D-CNN allows for the encoding of joined spatial-spectral features of a stacks of images. The 2D-CNN learns more extensive abstract spatial representative on top of the 3D-CNN. When compared to using a 3D-CNN alone, the usage of mixed CNN streamlines the model. By incorporating the 3D-dual attention mechanism module into the network, it can enhance the representation of image features in various dimensions, thereby increasing classification precision. Image features from multiple levels are represented in the image, improving classification accuracy. This article improves the classification accuracy of common datasets with the architecture DAMNet, while also validating the superiority of the attention mechanism on the Xuzhou dataset.

Zengzhao Sun, Huamei Xin, Bo Zhang, Jinwen Ren, Zhenye Luan, Hongzhen Li, Xingrun Wang, Jingjing Wang
Dual-Discriminator Based Multi-modal Medical Fusion

The common medical image fusion algorithms often have the problems of information redundancy and spatial information loss. Therefore, an adversarial network generation method based on double discrimination is proposed in this paper. Firstly, the dual discriminant network designed by us is used for the initial fusion, and then the fusion strategy is used to enhance the detailed features to complete the final fusion. The experimental results show that the fusion results obtained by our method have clearer spatial structure and more accurate spectral information compared with the common representative methods.

Haoran Wang, Zhen Hua, Jinjiang Li
Topology Design of Aircraft Cargo Loading Control System

With the continuous improvement of the national production level, the demand for air transportation is also increasing. The design of an aircraft cargo loading control system can meet the operation needs of modern high-speed logistics and greatly reduce the cost of air transportation. The design of communication network topology is an important part of the design scheme of aircraft cargo loading control systems. China’s air cargo has a late start, and has not formed a systematic research and development system in the aircraft cargo loading control system. For the self-developed cargo aircraft in China, the cargo loading control system is an indispensable part, in which the design and simulation of bus structure plays an important role. Aiming at the problem of communication network design of aircraft cargo hold loading control system, three kinds of network topologies are proposed in this paper. The proposed network topologies are based on typical network designs and general optimization indicators of the load rate and the real-time performance are taken into account. The feasibility of the proposed topologies is analyzed and compared using CANoe simulation.

Jie Deng, Chunyun Dong, Meng Nan, Siqing Wu, Xiaolong Chen
Information Bottleneck Attribution Based Retinal Disease Classification Using OCT Images

The existence of deep learning’s “black box” makes it difficult to understand how the algorithms analyze patterns and make image-level predictions. A representation of the pixels contributing the most to the algorithm’s classification will require new insights. Classification methods for neurodegenerative ocular disorders have been developed using machine learning and image processing techniques with considerable efficacy. However, the techniques’ robustness and transferability remain uncertain. We have developed a new classification method based upon information bottleneck to analyze the attribution of each feature and the information provided for the network prediction in each input area for a clear understanding of the affectability of a black-box model. In this research, we apply the attribution information bottleneck and limit the information flow and assess the amount of information image areas produce in bits by allowing noise to intermediate feature maps. Our studies indicate that the information bottleneck for attribution (IBA) has increased model interpretability and gives a more reliable estimation through a publicly available dataset.

Sehrish Aslam, Yuanjie Zheng, Xiaojie Li, Junxia Wang, Muhammad Zakir Ullah, Gogo Dauda Kaizolu, Neelam Gohar
Distributed Consensus for Discrete Multi-agent Systems with Time Delay in the Transmission

In our study, we analysed how to achieve consensus on the condition of distributed environment for a set of high-order discrete multi-agent models. Suppose the system is critically unstable and the undirected network which we analysed is under a time-varying communication delay. Through researching the union effects of network topology, communication delay and agent dynamics, we get the allowable delay bound which could be used to assure consensus. As a special case, if all poles of agent dynamics are located at the complex plane of $$z=1$$ z = 1 , consensus can be achieved for any bounded delay. Finally, a simulation result is presented to validate the feasibility of the obtained theoretical results.

Qipeng Hu, Yanli Zhu, Zhenhua Wang, Liming Zou
Simulation Design of Space Gravitational Wave Detection Satellite Formation

Gravitational wave detection is one of the hottest science topics in science today. However, due to the terrain conditions and the limitation of the detection arm length of the ground detection device, countries around the world have invested in the trend of space gravitational wave detection research. However, the cost of launching satellites is high, and the space environment is complex, which requires a lot of premise simulation experiments. This article aims to design a ground simulation method for space gravitational wave detection satellite formations, and a comprehensive scene display computer for simulation control and scene result display. The dynamics and component simulation computer simulates the space environment and the state of the satellite, and the attitude and orbit control computer verifies the control algorithm of the satellite. The design of this article provides a stable and effective simulation verification platform for space gravitational wave detection.

Wentao He, Xiande Wu, Yuheng Yang, Huibin Gao, Haiyue Yang, Yuqi Sun, Songjing Ma
Scheme Design for Cargo Loading Control System of Cargo Plane

Due to the late start of China’s civil aviation, the research on cargo loading control system of air freighter is less, and there is no research and development system and technical ability at present, so the research on related technology of loading control system is imminent. Aiming at the above problems, this paper studies the key technologies of CAN communication network design, software design and control hardware design of controller, and puts forward a design scheme of cargo loading system controller for large cargo aircraft. Firstly, based on the analysis of the composition of the loading control system and the requirements of loading and unloading, the basic function design requirements of the control system controller are proposed. Secondly, aiming at the characteristics of a large number of CAN communication nodes, a large amount of data transmission and high real-time requirements of the loading control system, a three-layer CAN communication network with sub-regional control is proposed to ensure the communication requirements and operation efficiency of the controller. On this basis, the hardware implementation scheme of dual MCU is proposed, and the application layer software design is based on an embedded real-time operating system. And the visualization function of PDU status and ULD real-time position is realized by using a serial display. The scheme proposed in this paper has a significant reference value for the automation and domestication of cargo loading control systems.

Siqing Wu, Chunyun Dong, Wenjuan Lei, Jie Deng, Xiaolong Chen
Chinese Word Segmentation of Ideological and Political Education Based on Unsupervised Learning

This paper proposes an unsupervised Chinese word segmentation algorithm for ideological and political education. The algorithm is divided into two parts: language model generation algorithm and Viterbi algorithm. Language model generation algorithm calculates the conditional probability based on the big texts and determines the number of occurrences between single character and character. Then we can have a character-level N-gram language model. Viterbi algorithm uses the idea of dynamic programming. Viterbi algorithm can use character-level language model to find the optimal word segmentation path, so as to further improve the accuracy of segmentation. Finally complete the task of Chinese word segmentation supported by big texts. Experiments show that the proposed algorithm has a good recognition rate for vocabulary in the field of ideological and political education. With the characteristics of unsupervised learning, the algorithm can save a lot of labor costs and meet the needs of word segmentation in the field of ideological and political education.

Xinghai Yang, Wenjing Zang, Shun Meng, Jiafeng Liu, Yulin Zhang
Random Design and Quadratic Programming Detection of Cayley Code in Large MIMO System

Cayley code can assure high rate transmission without channel estimation in Large multi-input and multi-output (MIMO) system, but it lacks systematic design and feasible detection method. In this paper, we propose a systematically random design and convex optimization detection scheme for Cayley codes with all kinds of data rate and all numbers of transmit and receive antennas. We employ complex Gaussian random matrices to generate Hermitian base matrices for signal design. Furthermore, we regard the transmission system as an equivalent large MIMO system whose transmit signals are the base coefficients, and then use quadratic programming (QP) for detection. In large MIMO system, when conventional non-coherent space-time codes are hard to design and the performance of conventional algorithms deteriorates, the proposed scheme can still operate. As the base matrices are randomly generated, our design has a high secrecy in nature.

Da-Peng Zhang, Fu-Yun Ma, Ting Liu, Ning Liu, Li-Jun Bian, Zhi-Min Zhang
The Design and Implementation of Mobile Operation Platform Based on Domain Driven Design

In order to actively promote the implementation of grass-roots burden reduction work of SGCC and improve the efficiency of power marketing field operation, this paper establishes a micro-application system for typical business scene by progressive six-step method, forms “building block” micro-application model for carrying all kinds of mobile operation business and application construction. On this basis, the business middle-ground quickly responded front-end micro-application for business changes and innovations is established by the domain driven design method, and constructs eight capability centers to support front-end micro-application. By this way, the architecture of “big platform, micro-application” is designed to support the fast iteration of new business, form shared micro service about common application and rule precipitation, and build the platform of “data feedback business” benign circle.

Wenjuan Shi, Shulong Wang, Jing Ma, Junquan Zhang, Yuxi Liu, Zaiyu Jiang, Yang Hu
Research on Ensemble Prediction Model of Apple Flowering Date

Apple flowering freeze is one of the major disasters affecting yield, and prediction of flowering date using meteorological factors is one of the important aids to reduce the impact of freeze damage. The prediction results of one single model are prone to fluctuations due to interannual and spatial changes. In this paper, 6 areas were selected from Shandong, Shaanxi, Henan, and Liaoning. By analyzing the meteorological data and apple flowering date in the last 10 years, key meteorological factors affecting flowering date were selected based on distance correlation coefficients. Later, the ensemble prediction model was constructed by using support vector machine regression, multiple linear regression, and decision tree regression as the base models. The results showed that the mean absolute deviation of the ensemble prediction model was in the range of 0.736–3.616, which showed good stability and prediction accuracy and could provide theoretical support for apple flowering date prediction.

Fan Zhang, Fenggang Sun, Zhijun Wang, Peng Lan
Apple Leaf Disease Identification Method Based on Improved YoloV5

In this paper, an improved YoloV5s based identification model, named as GSE_YoloV5s, is proposed for apple leaf disease identification, which can efficiently address the problem of high storage and computational resource consumption for the YoloV5s model. Firstly, the GhostBottleneck module is used to replace the original CSPBottleneck module to reduce the parameters and computation of the YoloV5s; meanwhile, by adding the channel attention module SE (Squeeze-and-Excitation), the model’s detection performance for small target lesions is improved. The experimental results show that, the improved GSE_YoloV5s model can reduce the number of parameters and computational effort by 40%, as compared with the YoloV5s model, and the AP (Average Precision) achieves 83.4%, which can effectively detect apple leaf diseases.

Yunlu Wang, Fenggang Sun, Zhijun Wang, Zhongchang Zhou, Peng Lan
Analysis of Apple Replant Disease Based on Three Machine Learning Methods

Apple replant disease (ARD) has become the limiting factor of the sustainable development of apple industry in China. The dry weight inhibition rate of apple tree rootstocks is one of the important indicators for judging the degree of apple replant disease. The calculation of dry weight inhibition rate needs to detect the dry weight of planted fruit tree rootstocks before and after soil sterilization. The specific operation is more troublesome. In response to this problem, this article proposes a method of using machine learning to predict the dry weight inhibition rate of fruit tree rootstocks. Using the existing measured data of the chemical composition, microbial community, and apple tree rhizosphere exudates in the soil in the orchard, the data is first preprocessed, including the division of the data set and the data normalization, and then the feature vectors are screened by the Distance Correlation Coefficient, and finally, using three models of Multiple Linear Regression, Support Vector Regression and Random Forest to predict the dry weight suppression rate respectively. The study found that the three machine learning methods are more suitable for the analysis of the experimental data. All three methods have obtained more accurate analysis and prediction results, and their R2_score is above 0.8. This study is of great significance for evaluating the severity of apple replant disease.

Guoen Chen, Li Xiang, Zhiquan Mao, Fenggang Sun, Peng Lan
Improved Federated Average Algorithm for Non-independent and Identically Distributed Data

Federated learning is a distributed machine learning model that can protect data privacy, where a large number of clients and servers participate in the joint train procedure. However, the problem of heterogeneous data has been a potential obstacle for federated learning. Traditional joint averaging algorithms suffer from low accuracy and slow convergence on non-IID (identical and independently distributed) data. To this end, this paper proposes an improved FedAvg based algorithm called FedAvg-Z. We introduce penalty coefficients and adaptive parameter averaging methods in our algorithm, which adjusts the influence of clients on the global model as a way to increase the convergence and stability of the model. We first demonstrate the reliability of our method on the dataset and then compare it experimentally with the traditional algorithm. The experiments show that the algorithm has higher accuracy and convergence speed in the non-IID case.

Zhongchang Zhou, Fenggang Sun, Peng Lan, Zhijun Wang
A Monitoring Scheme for Pine Wood Nematode Disease Tree Based on Deep Learning and Ground Monitoring

The detection of pine nematode trees in forest areas is a very important task to prevent and control pine nematode epidemic, where the timely and accurate detection in large and complex scenarios is a challenging work. To address this issue, we propose a ground-based monitoring detection scheme to acquire the large-scale images, and a single-stage deep neural network is applied to detect the infected tree. By installing monitoring cameras on the ground foundation, cameras are controlled to collect images automatically according to predetermined parameters. We designed an improved YOLOv3 based detection model, named as YOLO-S. Specifically, an attention mechanism is added to the backbone feature extraction network and the bottom-up feature pyramid structure is added to the original feature pyramid to enhance the detection accuracy. Experiments results show that the proposed YOLO-S algorithm can provide higher detection accuracy than that of other models.

Chengkai Ge, Fengdi Li, Fenggang Sun, Zhijun Wang, Peng Lan
An Optical Detection Method with Noise Suppression for Focal Panel Arrays

In practical engineering applications, due to the light reflection and scattering of the wireless channels with the suspended particles, such as the atmospheric channels or underwater channels, parts of the emitted optical signals will be reflected to the focal plane array of the space optical terminal. At the same time, due to the low isolation between the transmitting and receiving optical paths in the terminal, the reflected optical signal will interfere with the detection of the received optical signals from the far end, which seriously reduces the detection performance and the accuracy of the light spot centroid calculation. This paper proposes an increase-decrease statistical method for focal plane arrays, which can effectively eliminate the influence of the reflected optical signal and the noise on the signal detection. In addition, this increase-decrease statistical algorithm is simulated and verified by the two software co-simulation.

Bin Wang, Yuning Wang, Xiaoyan Wang
Orientation-Aware Vehicle Re-identification via Synthesis Data Orientation Regression

Owing to the need for smart city construction, vehicle re-identification (re-ID) has been widely used in the field of computer vision. Given a probe vehicle image, all of the same vehicles need to be found in the gallery data. However, because of the different camera shooting angles and vehicle driving directions, extreme changes in the viewing angle in vehicle images leads to a dissimilarity in shape, resulting in a difference in vision and having a significant impact on the accuracy. To address this issue, we propose a method for eliminating bias between different viewpoints, specifically, orientation regression training is conducted on a free synthetic dataset through the VehicleX engine, orientation-aware features are extracted using the trained network above, and the similarity calculated by the original re-ID feature and the orientation-aware feature are then fused to obtain the final similarity, which can effectively remove the orientation bias that exists with conventional re-ID features. Extensive experiments on two public datasets confirmed the effectiveness of the proposed method.

Hong Wang, Ziruo Sun
Opportunities and Challenges of Artificial Intelligence Education in Secondary Vocational Schools Under the Background of Emerging Engineering

In order to cope with the new generation of Scientific-technical revolution and industrial transformation, the construction of emerging engineering is in full swing. As the core technology of the new generation, artificial intelligence has had a disruptive impact on traditional industries and is an important part and module of the construction of emerging engineering. Secondary vocational schools are the cradle of cultivating applied technical talents and high-quality workers, the goals and content of talent training should closely follow the development trend of social technology and industry. However, artificial intelligence education in secondary vocational schools is still in the exploratory stage. Based on the development process of artificial intelligence education, this paper studies the opportunities and challenges of artificial intelligence education in secondary vocational schools under the background of emerging engineering, in order to provide a theoretical reference for the development of artificial intelligence education in secondary vocational schools in our country.

Ying Qu, Liancheng Xu, Ran Lu, Min Gao
Research on the Hotspots and Trends of Student Portrait Based on Citespace

In order to better play the role of student portrait technology in the field of education, this article displays and analyzes the research hotspot and research trend in this field. Taking the literature related to student portraits in CNKI as the research object, the author cooperation map, keyword clustering map and time line map are drawn by using Citespace. This paper analyzes the distribution of scientific research forces, research hotspots and research trends in this field. The study shows that student portrait is an emerging field of research, and there is a lack of cooperation between research scholars; Research hotspots in this field focus on big data, personalized education and student management and other topics; Smart campus, student behavior, big data, data mining and other information technologies represent the future research trends in the field of student portrait. Based on the research conclusions, this paper gives suggestions for future research on student portrait.

Jie Lv, Ran Lu, Zhe Liu
An Overview on Digital Content Watermarking

With the rapid development of multimedia technology, balancing multimedia visual enjoyment and information security has become one of the hot spots of current research. Due to the complexity and the cost of video content production, the corresponding video services face many security challenges, and higher requirements for the protection of video content copyright are also been put forward. Among them, digital watermarking technology is a typical information hiding method, which covers text, image and video. In the case of copyright disputes, watermark has become an important way to confirm the copyright and identity information. Digital watermarking technology uses the redundancy and randomness of media content data to embed the cipher text of copyright identification into the protected data. Except for data processing, the rest is invisible to ensure high security. This paper summarizes the current research progress of digital watermarking technology and digital watermarking algorithms, and then puts forward possible research directions in the future.

Wang Qi, Bei Yue, Chen Wangdu, Pan Xinghao, Cheng Zhipeng, Wang Shaokang, Wang Yizhao, Wang Chenwei
Location-Independent Human Activity Recognition Using WiFi Signal

The WiFi sensor constrains the Human activity recognition scheme to an immovable orientation when providing training samples, drastically reducing practical applications. In addition, the empirical investigation through the ubiquitous experience of participants in environments with wireless signals makes it imperatively challenging to evaluate a location-independent method to recognize human activity. In this research, a 1D CNN-LSTM model was built to comparatively analyze location-independent human activity recognition through WiFi CSI. Thus, the method reduces the impact of human activity recognition on location independent. Nevertheless, the experimental results demonstrate that our method can achieve over 94.9% coverage accuracy for location-independent human activity recognition and 90% coverage accuracy for CNNs using our proposed method. The location-independent model developed can be used possibly for practical applications in any situation, and data from each antenna is processed independently, making it applicable in any location or antenna setup. A variety of input sample lengths has been tested to overcome the measurement sensors’ sampling rate limitations. The results show that human activities can be recognized in real-time through WiFi signaling. Therefore, the CSI WiFi connectivity is potentially enhancive in building an excellent platform to locate human activities in modern age.

Gogo Dauda Kiazolu, Sehrish Aslam, Muhammad Zakir Ullah, Mingda Han, Sonkarlay J. Y. Weamie, Robert H. B. Miller
Design of Text Summarization System Based on Finetuned-BERT and TF-IDF

In today’s world, information is exploding. Internet readers need to judge the readability of the text by reading abstracts when searching for some information, such as blogs and technology papers, but the current text information is generally lack of abstracts. The development of NLP (Natural Language Processing) technology has a great help to fill the blanks. The purpose of this paper is to propose a text summary system, which has been tested in the corpus (including CLTS and some educational papers) and shown great results in the shorter samples. Our text summary system provides two options for the text without abstracts or with poor original abstracts, which are based on finetune-BERT and TF-IDF respectively.These two strategies are different from processing speed and semantic fit. In addition, in order to apply to more diverse scenarios, this article makes it possible to generate a summary of the image text by incorporating the OCR (Optical Character Recognition) technology.

Yuyang Liu, Kevin Song, Jin He, Songlin Sun, Rui Liu
Rogue WiFi Detection Based on CSI Multi-feature Fusion

At present, using WiFi to access the Internet has become a mainstream way. However, through research on wireless network security, it is found that attackers forge the mac address or SSID of legitimate users to obtain illegal information. This method is difficult to be detected by traditional network security mechanisms. In this article, we propose a new security mechanism that uses multiple features extracted from Channel State Information (CSI) which are irrelevant to the environment for fusion, and uses a minimum discriminant criterion for classification and discrimination to reject rogue WiFi connections. We found that the Carrier Frequency Offset (CFO) and nonlinear phase error extracted from CSI are stable in time and space. Experiments on a large number of WiFi devices show that this mechanism can reliably detect rogue WiFi connections, with an average recognition rate over 96%. At the same time, through comparing the recognition effect between before and after the feature fusion, we verify the superiority of the fusion method, and provides a reliable basis for the identification of rogue WiFi.

Zhaoyuan Mei, Songlin Sun, Chenwei Wang
Panoramic Video Quality Assessment Based on Spatial-Temporal Convolutional Neural Networks

The development of 5G technology and Ultra HD video provide the basis for panoramic video, namely virtual reality (VR). At present, the traditional VQA method is not effective on panoramic video. Therefore, it is crucial to design objective VQA models for the standardization of panoramic video industry. With the development of deep learning, excellent algorithms of VQA methods based on convolutional neural network have emerged. In this paper, we propose a full reference VQA model based on spatial-temporal 3D convolutional neural network, the feature extraction combined the time and spatial information. we verify and optimize the proposed VQA model based on VQA-ODV panoramic video database, its objective score has a higher correlation with subjective scores than that of traditional VQA methods.

Tingting An, Songlin Sun, Rui Liu
Fast CU Partition Algorithm for AVS3 Based on DCT Coefficients

Audio Video coding Standard 3 (AVS3), which has been brought in many advanced algorithms and technologies on the basis of previous video coding standard, has already improved the coding efficiency greatly. However, with the performance improvement comes a huge increase in coding time. To facilitate commercialization, algorithms that reduce the time complexity of AVS3 standard must be developed. Based on the new code block (CU) partitioning rules of AVS3 standard, we present a new CU partition algorithm based on the discrete cosine transform (DCT) coefficients. According to the feature that DCT coefficient can reflect image texture, we find the relationship between it and CU partition, The DCT coefficients calculated by the fast DCT algorithm are used to limit and make decisions on CU partitioning in all-intra (AI) mode, so that the ergodic process of CU partitioning is greatly reduced, thus achieving the effect of reducing the time complexity. Experimental results show that the proposed algorithm can reduce the time complexity with a small Bjøntegaard-Delta rate (BD-rate) increase.

Xi Tao, Lei Chen, Fuyun Kang, Chenggang Xu, Zihao Liu
Research on Recognition Method of Maneuver and Formation Tactics in BVR Cooperative Combat Based on Dynamic Bayesian Network

With breakthroughs in modern weaponry and equipment technology and coordinated formation flying, coordinated operations beyond visual range have become the main form of modern air combat. Accurately identifying enemy maneuvers and formation tactics is of great significance for grasping the dominance of the battlefield and grasping the direction of war. Traditional situation assessment research lacks pertinence to coordinated formation tactics and strategies in the BVR battlefield. This paper considers the influence of the characteristics of BVR fighters on tactical strategies, analyzes the causal relationship between the characteristics of flight parameters and the evolution of tactical movements, and constructs a dynamic Recognition model of typical maneuver movements beyond the visual range based on Bayesian network; on this basis, a recognition model of cooperative formation tactics is constructed. Finally, simulation experiments verify the accuracy and effectiveness of the model’s recognition. Further, this model can be used as a threat module of the beyond-horizon cooperative formation tactics for the Air combat situation assessment.

Yan Zhang, Yuyang Liu, Yinjing Guo, Xiaodong Wang, Lei Feng, Guo Li, Yangming Guo
Fast Intra Prediction Algorithm for AVS3

With the development of 5G and multimedia technologies, applications such as ultra-high-definition video, VR and High dynamic range video have become more popular. This has led to an increasing amount of video data, which has brought great challenges to the storage and transmission of video data. Therefore, a video encoder with better compression performance is needed to remove the time redundancy, spatial redundancy and statistical redundancy in the image sequence of video, and greatly compress the raw video data under the premise of ensuring image quality. The Audio Video Coding Standard Workgroup of China has formulated the AVS video coding standard, and the latest AVS3 standard is currently under research and formulation. In order to improve the compression performance of the encoder, AVS3 introduced a large number of new technologies on the basis of AVS2 standard, resulting in a 266% increase in encoding complexity compared to AVS2. This paper optimizes the coding unit partition process and intra mode decision process of AVS3 Results show that the proposed algorithms reduce about 46.82% and 4.49% intra encoding time on average, with only 1.22% and 0.01% increases in terms of BDBR.

Lingfeng Fang, Songlin Sun, Rui Liu

Big Data Workshop

Frontmatter
Research on Interference Source Identification and Location Based on Big Data and Deep Learning

The thesis is mayor in based on the evaluation of base station interference sources in multiple dimensions by China Unicom Shanghai Branch. Based on the value mining of multiple data sources such as KPI/XDR/MDT, it closely integrates the traditional interference source investigation process and big data/machine learning Technology. At the same time, combined with BPNN neural network, the interference source and interference type can be identified intelligently. By using the BFL boundary point fitting algorithm, the interference source can be detected, classified and simulated. At the same time, a set of interference source overall optimization scheme is given.

Hui Pan, ChengFan Zhang, Saibin Yao, JiuCheng Huang, YiDing Chen, Long Zhang
Research on Automated Test and Analysis System for 5G VoNR Service

Automated testing technology can effectively save human resources, material resources and time costs. The introduction of automated testing technology into the communication network testing can achieve accurate testing of designated networks and designated functions. To ensure the user perception of the 5G VoNR service, an inherent advantage of the operator, under the condition of rapid iteration and frequent upgrades of the 5G cloud core network, an automated test and analysis system for 5G VoNR service is studied in this article. This system has three primary functions: simulation dial tests, quality evaluation and automatic early warning. Finally, the application results of this system in the live network are introduced, which proves its practicability.

Zhenqiao Zhao, Xiqing Liu, Jie Miao, Shiyu Zhou, Xinzhou Cheng, Lexi Xu, Xin He, Shuohan Liu
Mobile Network Monitoring Technology for City Freeway Based on Big Data Analysis

In the scene of city freeway, the driving speed is relatively fast, the flow of people is denser, and the customers have higher requirements for communication experience. Firstly, the architecture of big data platform for mobile network quality control was introduced. Then, the mobile network monitoring scheme for city freeway based on big data analysis was proposed. The scheme is oriented to group perception which approaches real experience, so as to effectively improving the evaluation ability and frequency for mobile network of city freeway. It automatically identifies users who driving on freeway from the data lake of XDR by clustering algorithm. Through 3W matching association, the signaling data pool of freeway users was generated, which would contribute to reduce the data volume and the consumption of computing resources greatly. Through road section analysis of freeway user trajectory, KPI monitoring and warning of abnormal for road section were realized.

Baoyou Wang, Saibin Yao, Jiucheng Huang, Hui Pan, Chao Liu
5G Uplink Coverage Enhancement Based on Coordinating NR TDD and NR FDD

With the development of the 5G vertical industry, significantly improving uplink experience and shortening latency without changing downlink experience is a new requirement and challenge for networks. The uplink coverage of the 5G network is different from that of the downlink coverage due to different antenna quantity settings, meanwhile, the 5G network can better compensate for the problem of large loss caused by high frequency bands on the base station side. As a result, the downlink coverage performance is far greater than the uplink coverage in 5G network. This paper proposes an uplink and downlink decoupling SUL technology for enhancing uplink coverage performance of 5G network and an uplink enhancement technology for coordinating NR TDD and NR FDD so as to uplink data is transmitted on the SUL spectrum or the NR FDD spectrum by time-sharing of the uplink data. This greatly increases the available uplink time-frequency resources for 5G UEs.

Jinge Guo, Yang Zhang, Bao Guo, Zhongsheng Fan, Huangtao Song
Application of Improved GRNN Algorithm for Task Man-Hours Prediction in Metro Project

In order to improve the accuracy and efficiency of subway project task man-hours prediction, a generalized regression neural network (GRNN) prediction model based on the improved Sparrow Search Algorithm is proposed. And a prediction study was carried out on the task man-hours of subway projects, and the factors affecting the design task man-hours were analyzed by taking the architectural profession as an example. The algorithm of this paper was used to predict. The results show that the improved GRNN algorithm proposed in this paper has a smaller error and a higher accuracy, and it has great potential for application in the prediction of the task man-hours of subway projects.

Zhengyu Zhang, Shuying Wang, Jianlin Fu
Resident Area Determination Model Based on Big Data and Large Scale Matrix Calculation

User’s resident area is very important for mobile network planning and customer maintenance. But the traditional methods which is always based on traffic calculation without consideration of neighbor cells or user’s spatio-temporal trace are not sufficient for now. In this paper we present a brand new method based on wide sources of data include traffic usage record and signaling record, meanwhile big data processing and matrix calculation are introduced. All the traffic usage record and signaling record can describe the user’s trace in contrast to traditional method in detail. The determined resident area could be one or a cluster of cells. With some experiment test, this method has got effective and accurate result. The method is used for a cell level location determination and finding a balance between the calculation complexity, resource cost and real-time performance. And a big data calculation module is under developed in one operator’s data middle ground in use.

Chuntao Song, Yong Wang, Xinzhou Cheng, Lexi Xu
Research of New Energy Vehicles User Portrait and Vehicle-Owner Matching Model Based on Multi-source Big Data

Due to the national policy of energy conservation and emission reduction, the new energy vehicular industry experiences fast development in China, and the sales of new energy vehicles have grown rapidly in the past two years. China Unicom provides new energy vehicles with the gamut of network services and has collected large volume of network interaction data with potential values. By constructing a model system for mining potential information value, this paper proposes innovative methods of dissecting the characteristics of new energy vehicles’ consumers and building user portraits, as well as matching the relationship between car owners and vehicles based on multi-source big data. This will effectively assist the precision marketing of the new energy vehicular industry and the landing of innovative cooperation scenarios between China Unicom and the new energy vehicular industry.

Yuting Zheng, Tao Zhang, Guanghai Liu, Yi Li, Chen Cheng, Lexi Xu, Xinzhou Cheng, Xiaodong Cao
Cell Expansion Priority Recommendation Based on Prophet Algorithm

In recent years, both LTE and 5G experiences fast development and construction world-widely. Under the explosive business demand and massive mobile terminal equipment connection, the guarantee of network load is of vital importance. This paper proposes a recommendation model for cell expansion priority. Initially, it uses the Prophet algorithm to predict cell traffic trends based on user behavior characteristics in different scenarios. Then it defines expansion warning thresholds according to different expansion types. Finally, combining cell traffic trends and expansion thresholds, a recommendation model is established. The proposed model can assist network optimizer more accurately grasp the current network capacity situation and future capacity trend, monitor the indexes of network expansion dynamically. Besides, this model can also follow the expansion principle of timeliness and predictability, which can make rational use of expansion investment and guarantee user perception.

Xiaomeng Zhu, Zijing Yang, Guanghai Liu, Yi Li, Lexi Xu, Xinzhou Cheng, Runsha Dong
Load and Energy Balance Strategy of SDN Control Plane in Mobility Scenarios

In this paper, We propose a solution for dynamic placement of controllers in software-defined networking (SDN), with an aim to solve the problem of unbalanced load and energy in SDN control plane. SDN architecture separates control and forwarding of network, provides the advantages of network programmable and easy expansion, and is of great significance in the application of the Internet of Things. Due to the dynamics of network load and device location, SDN control plane faces the problem of unbalanced load and energy. The load of some controllers is significantly higher than other nodes, resulting in a decrease in network performance; in a scenario where the energy of the control node is limited, some key nodes may run out of energy faster and shorten the life of the network. The dynamic placement of the controller is an effective way to solve this problem. This study uses the Oder-3 Markov predictor to predict the future access point location of the device, then used the branch and bound method to dynamically place the controller. The simulation results show that this method can effectively reduce the peak traffic of the SDN control plane and balance the historical load value of nodes.

Fuwei Zhang, Wenwen Ma, Dequan Xiao, Hongyu Peng
A Method for Intelligently Evaluation the Integrity of 5G MR Data

Measurement Report (MR) comes from all actual user equipment that use the service, it has the characteristics of comprehensive data and authentic source, which has been widely used in network optimization. This paper proposes a method for intelligently evaluating the integrity of 5G MR data based on machine learning algorithms, to get the standard sample data by modeling the number of MRO sampling and the relationship function of wireless performance, which verifies the integrity of the MR data reported on the existing network, thereby effectively improving the MR data quality.

Yuan Fang, Qintian Wang, Ao Shen, Pengcheng Liu, Jinhu Shen, Bao Guo, Zetao Xu, Jimin Ling
Internet of Vehicles Product Recommendation Algorithm and Implementation Based on Ensemble Learning

With the development of 5G, big data and other new technologies, the Internet of Vehicles is recognized as the future development direction of the industry. The traffic of operators ensures the network connection of vehicles, but how to recommend traffic products for the appropriate users has always been one of the research topics of operators. This paper implements a recommendation system based on big data of operators. The system can mine users’ behavior characteristics and use machine learning method to predict the possibility of users accepting the recommended products. In order to verify the effect of the model recommendation, this paper compares the number of orders and flow growth rate before and after using the model, and the results show that the model-based user payment rate is significantly higher than those that do not use the model.

Mingjing Zhang, Fei Su, Yanru Li, Guangtao Zhou, Liang Xin, Tingting Ren
Research and Verification of Power Saving Technology in 5G Network

With the development of 5G networks, the scale of 5G base stations is rapidly expanding, and the energy consumption of equipment is increasing rapidly. This paper introduces several existing wireless power saving technologies for 5G base stations, and then uses various technologies to carry out single-station power saving tests in the pilot area. By comparing the differences in base station energy consumption and network performance indicators before and after the implementation of power saving measures, the impact of this technology on power saving and network performance is verified.

Zetao Xu, Yuan Fang, Ao Shen, Yang Zhang, Pengcheng Liu, Jinhu Shen, Jimin Ling, Qintian Wang
Application of XGBOOST Model on Potential 5G Mobile Users Forecast

As the 5G mobile communication started the business services from 2019, a lot of mobile companies began to make mobile phones suitable for 5G internet. However, not all users have changed to 5G internet supportive mobiles. Therefore, Chinese operators are interested in identifying what kind of users are target 5G customers in order to promote the sales of them. Despite unsupervised learning models performed well in predicting customer behavior, we were interested in employing supervised learning model, XGBOOST model, to provide solid prediction. Besides, we were also interested in identifying the features and source-systems, which contribute more to the prediction. In this paper, we use XGBOOST model on data both from operation support system and business support system to predict future 5G mobile users. Results show that the XGBOOST model reaches the accuracy of nearly 70% in classifying 5G mobile users in general. In addition, streaming data from operation support system tended to be more important in predicting models than that from business support system.

Xijuan Liu, Tianyi Wang
Research on Methodology of Symmetry Optimization with Backpropagation

Machine learning is a promising technology, which is widely employed to structure a rather accurate physical system. ML-assisted application in the field of elementary particle becomes a hot-topic in both academics and industries. Electroweak interactions in the context of Standard Model is heavily depend on the Higgs mechanism. When spontaneous symmetry breaking happens, the challenge is to find a way to describe the theory behind it. In order to address this challenge, this paper investigates the breaking of global and local −U(1) symmetry and provides a pathway to compute the optimization process of nature system with gradient descent. Initially, this paper introduces Lagrangian to describe the status of a physical system which provides the initial data set. Then, this paper constructs symmetries and discovers the Higgs mass, which relates to the breaking of U(1) symmetry. Furthermore, trained network can be used to illustrate the loss function and give both learning rate and feature of the physical system.

Sitong Zhang, Youli Yu, Chenyuan Ye
Backmatter
Metadata
Title
Signal and Information Processing, Networking and Computers
Editors
Prof. Jiande Sun
Dr. Yue Wang
Mengyao Huo
Dr. Lexi Xu
Copyright Year
2023
Publisher
Springer Nature Singapore
Electronic ISBN
978-981-19-3387-5
Print ISBN
978-981-19-3386-8
DOI
https://doi.org/10.1007/978-981-19-3387-5