Signal and Information Processing, Networking and Computers

Compared to Low Earth Orbit (LEO) synthetic aperture radar (SAR) satellites, geosynchronous orbit (GEO) SAR satellites have the advantages of wide swath, fast revisit, and continuous observation. Due to the influences of Earth rotation and satellite orbit parameters, such as inclination and eccentricity, the Doppler center frequency of the satellite borne SAR imaging is always non-zero while the attitudes of three axis remains nearly zero, resulting in degraded SAR image retrieval quality. The Yaw-Pitch two-dimensional attitude guidance strategy proposed by existing researches can achieve zero Doppler center frequency. But when the Yaw-Pitch guidance strategy applied to GEO satellites, the yaw angle significantly varies. Take the GEO orbit with 16° inclination for example, the yaw angle fluctuations from −82° to + 82°, causing great engineering design problems especially in energy, layout, etc. This paper proposes a Pitch-Roll two-dimensional attitude guidance strategy, which also achieves zero Doppler center frequency by realizing the same target pointing as the Yaw-Pitch guidance strategy. The pitch and roll angles variation range achieved by this method is much smaller, which is more favorable for engineering implementation.

The multi-instruction execution function is a necessary function for the use of spacecraft on-board equipment. In this paper, based on the traditional multi-instruction execution strategy, an on-board complicated multi-instruction parallel processing technique is proposed, which can make the multi-instructions more flexible and agile in use, reduce the on-orbit bus transfer occupancy rate, enable the formulator of multi-instructions to abstract a variety of generalized templates, and these templates to make the system suitable for a variety of modes. The on-board software opens multiple buffers that hold multiple sequences of single instructions registered for multi-instructions, and finds and executes whether there is a single instruction that needs to be executed in each registered multi-instruction, so as to achieve the purpose of parallel execution of multi-instructions. Experiments show that this technique can make the use of multi-instructions more flexible and agile, reduce the bus transmission occupancy, the maximum ratio of execution time deviation to interval is 3.75‰ and prove the feasibility of this technique. This technique has guiding significance for the execution of multi-instructions in complicated on-board control software.

Traditional aerospace equipment development is highly integrated with software and hardware, which is not conducive to hardware sharing and software reuse, and on-orbit reconfiguration capability is limited, severely restricting the strategic development of the aerospace industry. The article proposes the architecture of software-defined highly credible space equipment, focusing on solving the pain points existing in the traditional development mode, aiming to realize the decoupling of software and hardness through the establishment of hardware abstraction layer by embedded real-time operating system, and verifying its credible mechanism through formal modeling, standardize software programming, clarify logic and automate testing with portability and deterministic.Software-defined technology is expected to fundamentally change the production mode and industrial ecology of spacecraft, triggering a new round of technological innovation in the aerospace industry.

Intellectual property is an important carrier of innovation protection and a powerful tool for modern technology-intensive enterprises to compete and cooperate. As we all know, applying for patents is the preferred way for enterprises and scientific institutes in various countries to protect advanced technologies, and patent literature records more than 95% of the world’s innovation achievements, which is an indispensable source of technology for scientific researchers to carry out technical research. Through patent literature retrieval and analysis, the development trend of a certain technology can be known. This paper analyzes the patent application situation, right holder distribution, technical topics, technical efficacy and technical hot spots of space-borne computers, to understand the development trend of space-borne computer related technologies and the global intellectual property distribution. Through the research of this paper, we know that the field of spaceborne computer began to apply for patents as early as the 1970s, and the United States aerospace enterprises have mastered the core key technologies in this field. In recent years, China’s patent layout in this field has continued to grow. The research results of this paper can clarify the technical route and fill the technical gap for the follow-up research on spaceborne computing, and provide important hot field technical solutions for technicians as a reference to improve the starting point of research and development.

With the depth integration of new generation of information technology and space technology, the satellite system will undergo subversive changes and is developing in the direction of systematization, networking and intelligence. The thesis analyzes the development status of intelligent satellite network system, analyzes the development needs to accelerate the system construction, intelligent application and manufacturing mode transformation, and discusses the general characteristics of the new-generation intelligent satellite network system, which is transforming from a single satellite to a satellite system, from acquiring data to acquiring knowledge, and from a general electromechanical system to an intelligent system. The paper proposes an intelligent information system centered on high-performance computers, universal operating systems and intelligent application algorithms, studies key technologies such as task planning, intelligent processing and network interconnection for accelerating the development of the intelligent satellite network system, and gives the medium- and long-term development routes, key indexes and general recommendations for the new generation intelligent satellite network system.

The design of efficient and fast autonomous task planning algorithms is the core of achieving satellite high maneuverability autonomous task planning with multiple boundary constraints. This article studies a satellite high maneuverability autonomous task planning algorithm with multiple boundary constraints, using a distributed method to allocate and execute observation tasks released on the ground independently, reducing the pressure on ground personnel to control satellites, and achieving full utilization of satellite observation resources. After receiving various tasks, satellites can independently complete planning tasks such as regional target segmentation, multi-objective timing arrangement, task execution/switching process planning, and task conflict selection based on environmental information and their own state information. This article simulates and verifies the algorithm through typical application scenarios, demonstrating its effectiveness.

Bamboo has the advantages of short growth cycle and fast maturity, it spends only 3–5 years to become forests. From the perspective of carbon sequestration processing, bamboo products can be completely degraded naturally after use, bamboo resources have great potential and prospects in solving the global plastic problem and reducing pollution emissions, and are an indispensable part of the “Carbon emission peak and carbon neutrality strategic goal”. In this paper, a satellite constellation design method was proposed to monitor the biomass carbon sink and calculate GEP based on the background of bamboo forest resource monitoring in China, combined with visible light, hyper spectrum and synthetic aperture radar (SAR) loads.

The imaging mechanisms of SAR sensors and optical ones are different. Accordingly, there are differences between features of SAR and optical images. The fusion methods for SAR satellite images and optical ones mainly include wavelet transform, HIS transform, PCA transform, and so on. Each transformation method has different performances on their fused images. In the paper, the performance evaluation method of SAR and optical image fusion methods based on real-world satellite images is studied. Firstly, SAR and optical image fusion process is introduced, including image preprocessing, image registration and image fusion. Secondly, various evaluation indicator, such as average gradient (AG), distortion, entropy, mutual information (MI), universal image quality index (UIQI) and edge mutual information (EMI) are provided. Finally, taking Yangzhong Island region in China as an experimental site, multiple image fusion methods were used to fuse a GF-3 satellite SAR image and an optical one. Evaluation criteria are provided for the applicability of image fusion methods.

As we enter a new stage of aerospace development, traditional aerospace project evaluation models can no longer meet the needs of aerospace enterprises. The era of big data provides technical support for the systematic, model-based, and intelligent management of aerospace enterprise project evaluations. Based on big data layered processing and using information technology as a tool, this paper proposes a project evaluation system for aerospace enterprises, which includes three levels DOS, DGL and DSL. It also explores an integrated data evaluation model, which can reflect the value of different projects to enterprise development in a real-time, accurate, and visual manner. This system collects various types of raw data to the big data platform, manages the original big data based on the data model; and provides effective guidance for optimizing management mechanisms in aerospace enterprises. We hope it can generate a powerful synergy for innovation and support high-quality development of aerospace enterprises.

Spacecraft charging to the magnitude of kV may be harmful to the health of onboard electronics. New equivalent circuit model is adopted in the surface charging analysis in this paper. Voltage controlled current source (VCCS) is used to simulate the injected current of hot plasma instead of the popular used constant current source. The equivalent circuit model was built to calculate the charging potential of each node. The model is performed under two conditions, one is spacecraft in shadow and the other is in sunlight. At the same time, in order to verify the rationality and accuracy of the equivalent circuit model, taking the satellite in the shadow area as an example, the charging equivalent circuit of the solar cell array in the plasma environment is established, and the potential values of the structure and the solar cell array are obtained.

This thesis summarizes the construction of a digital financial management platform in the operational management process of aerospace enterprises, covering comprehensive budgeting, contract management, and full life-cycle funding for projects, among other financial management tasks. By establishing a digital financial platform based on unified data sources, conducting in-depth data governance, and achieving interconnection and sharing of financial and business information, a management dashboard based on real-time data is created. Cost data is accurately delivered to different users through cost models, ensuring timely dissemination of financial information. By assetizing financial data and modeling business processes, standard tools and methods are provided for financial management in aerospace enterprises, ensuring that costs are known, visible, and controllable, and effectively improving the level of financial meticulous management.

With the rapid development of commercial aerospace, the number of satellites is increasing every year. While focusing on satellite manufacturing, the demand for satellite on-orbit operation management is also increasing. To meet the increasing requirements of satellite on-orbit operation management, this paper presents a design of FCS (Flight Control System) for satellite constellation. Starting from the system architecture, it describes the front-end design of the ground station, the design of the cloud data center and the design of user center, and gives the telemetry and telecommand data flow. Using the proposed system design in this paper, the full link and high-efficiency real-time processing of satellite on-orbit management can be realized, which is more in line with the development trend of commercial satellite industry.

With the establishment of the Beidou Navigation System in China, RGNSS becomes an ideal relative measurement device for space rendezvous and docking missions. In theory, it can meet the relative measurement needs of most cooperative types of space rendezvous and docking missions under full range conditions. Especially for the applications of micro-satellites, RGNSS has a great help in addressing resource constraints such as weight and size. However, due to the significant impact of external factors on data stability, the use of RGNSS in close range conditions has become a challenge in system design. Based on summarizing the principle of RGNSS carrier phase difference and carrier phase ambiguity error, this article proposes a system level RGNSS usage strategy that can eliminate the impact of RGNSS integer ambiguity errors on spacecraft navigation stability and achieve the goal of using RGNSS at close range conditions.

To the issue of far-field safety of multi-body in multi-satellite launch mission with same rocket, this paper presents the influence relationship between different separation velocities and orbital parameters based on orbital dynamics, and presents the design requirements for separation velocities that meet far-field safety constraints. And then a prediction and review method of multi-body flight spacing is provided based on the worst-case force analysis model, then a complete set of far-field safety analysis and design method is proposed for multi-satellite launch mission. The implementation results of actual flight missions indicate that this method can effectively ensure the far-field safety of multi-body after satellite rocket separation, and is helpful in guiding the safety design of multi-body separation for multi-satellite launch mission with same rocket.

This paper provides a detailed introduction to the key technology and software design of Failure warning system of floated gyro based on mechanism; The software is developed based on SpringCloud Microservice architecture. The fault warning model is constructed according to the floated gyro fault mechanism, statistical analysis method and trend prediction algorithm design. The state identification and warning of the floated gyro is achieved by receiving real-time telemetry data or simulation data. For the first time, the sub-health state is defined between the health state and the fault state, and the diagnosis criterion is designed according to the model principle. The on-orbit actual test case and the 1000-h actual system verification prove that the early detection of the sub-health state is conducive to the health monitoring and on-track maintenance of the floated gyro while on orbit. It can realize the forecast and warning of typical faults with a high accuracy, and expand the fault alarm and prediction suitable for other key on-orbit single machines.

In the process of spacecraft development, schedule management involves numerous influencing factors, which constitute the risk of project schedule management from a risk perspective. This paper analyzes the new characteristics of spacecraft project schedule management, explores and improves the identification and evaluation methods of project schedule risks, and further studies the level evaluation of technical implementation risks, organizational management risks, and environmental change risks in different development stages of spacecraft projects. Corresponding spacecraft project schedule risk management strategies are proposed.

To realize high-speed and reliable transmission of navigation information data and processing results to other devices. This paper constructs TCP hardware protocol stack based on FPGA. According to the idea of hierarchical processing and modular design, the complex data transmission process is simplified into the process of data “encapsulation - cycle - unencapsulation”. The functions of address resolution protocol ARP, TCP connection establishment, data communication and disconnection are successfully implemented. Test the TCP transmission rate. When the data transfer is stable, the maximum speed can reach 379 Mb/s. It basically meets the network transmission requirements of satellite navigation equipment at the present stage. In addition, this paper also realizes the TCP/IP protocol in a pure logic way. If it can be integrated into the chip through the complete IC design verification, the prospect is incalculable.

In high-resolution remote sensing satellites, the cable stiffness has a significant impact on the vibration-isolation effect and stability. To obtain the accurate stiffness characteristics of the on-board cable, the stiffness experimental testing system was built based on the substructure stiffness equivalent method. The paper analyzed stiffness effect factors according to the experimental data and put forward feasible methods to weaken the additional stiffness. Finally, the Direct Matrix Input on Grid Points method was used to compile the pre-processing file by inputting stiffness matrix information. The cable finite element model is set up to output the total stiffness matrix, which realize the simulation and integration of the cable group.

In the process of attitude maneuver, whether Single Gimbal Control Moment Gyroscopes (SGCMGs) can provide control torque accurately is the key of spacecraft attitude maneuver mission. In order to solve the problem that the system easily fall into singularity and saturation state caused by the complex configuration of SGCMGs, the gimbal angle motion based on stable interval search (GMSS) planning method is proposed. The SGCMGs gimbal angle motion planning model is established to transform the SGCMGs path solving problem into constraint satisfaction problem. Combined with the defined singular measure and saturation measure, the motion strategy of SGCMGs in gimbal angle space is designed and the singular constraint and saturation constraint are transformed into path constraint in gimbal angle space. By using the stable interval search strategy, the angular momentum solution space of SGCMGs system is expanded, and the angular motion path of SGCMGs gimbal needed for attitude maneuver can be predicted. Finally, the simulation results show that the proposed method can generate the angular motion path of SGCMGs satisfying the constraints.

The target capture of space multi-arm robot is one of the key technologies that need to be solved in the engineering of in-orbit service in the future. Multi-arm space robot is an integrity constraint, and there is dynamic coupling and interference between its base and mechanical arm. This disturbance exerted by the mechanical arm to the movement of the base spacecraft will affect the working accuracy of the space robot to some extent, and will lead to the failure of the in-orbit operation mission. In this paper, the kinematics and dynamics model of the spatial two-arm robot is established, the capture path algorithm is designed by the proportional guidance method of the operating arm, and the motion planning of the compensation arm is designed to eliminate the disturbance of the translation and rotation of the base by the operating arm. The motion simulation is carried out on the basis of the example.

To improve the spatial resolution of remote sensing images, an improved Projections onto Convex Sets (POCS) super-resolution algorithm based on Point Spread Function (PSF) estimation is proposed. Obtaining PSF from system design or remote sensing image, we use it as a projection operator to simulate the real degradation process of the satellite imaging system, and improve the reconstruction accuracy. To evaluate its performance, we compared it with the classic POCS algorithm, the PSNR/SSIM of the super-resolution image reconstructed by our algorithm is improved from 24.32dB/0.77 to 25.73dB/0.82, and the visual effect of the image is significantly enhanced.

Aiming at the real-time on-board intelligent decoding/translating need of the satellite remote sensing image, the article raises up a system structure of satellite remote sensing image terrain classification based on the combination of terrain servers and on-board AI analyzing platform, and designs an on-board satellite remote sensing image terrain classification algorithm based on deepabv3+. Then, the optimizing solution satisfies the constraints of the performance and the storage of the on-board intelligent information system is offered. Tested by on-board AI analyzing platform built on land, the solution is proved to be able to satisfy the on-board real-time processing need, the on-board image decoding/translating speed and accuracy requirements.

The development process of spacecraft AIT has the characteristics of complex processes, If there is a slight carelessness, it will have an unpredictable impact on the quality, progress, and cost of spacecraft development. Therefore, it is necessary to focus on strengthening the risk identification and control of product development, particularly strengthening the dynamic control of risk monitoring links. This article describes the methods and practical application effects of dynamic risk management for spacecraft technology, and proposes the direction and content that need to be improved for future technological risks.

Spacecraft flight effectiveness evaluation is an indispensable and important link in whole-stage test verification and on-orbit flight result evaluation. In allusion to the deep-space high-speed saltatory reentry and return flight test, an evaluation scheme of spacecraft flight effectiveness based on ADC model is proposed. In the study, the flight mission profiles are divided to establish the index evaluation system facing deep-space reentry and return process while the ADC effectiveness model is analyzed and calculated. Meanwhile, by virtue of the effectiveness quantification function, the capability metric of each mission profile is calculated, thus solving the tricky problem that the system target completion capability is not easily described quantitatively, and realizing the scientific quantitative assessment of autonomous flight capability in the navigation courses such as the establishment section, the reentry section and the leaping section, which is of certain reference significance for the subsequent effectiveness evaluation of the probe systems such as unmanned Mars or small celestial body sampling return.

The technology of products in the space industry is advancing in rapid pace, resulting in prices variation. The purpose of space product price prediction is to make scientific estimations and speculations on the trend of space product prices based on specific methodology, survey, and mathematical analysis, with a focus on the available statistical data and relevant information to unveil the correlations that underlay procurement activities. Efforts have been made in prediction methodologies taxonomy, qualitative and quantitative prediction, methodology development, data modelling establishment and correction, which provide concrete theoretical support for the development of aerospace products and price prediction.

In recent years, indoor positioning services have been widely used in the fields of navigation, emergency rescue and disaster management. Among various indoor positioning technologies, indoor positioning based on Bluetooth sensor is easy to be popularized and has strong practicability, moreover, it can be realized with the assistance of some built-in functions of mobile terminals. However, the positioning effect is often poor in areas with many people and complex electromagnetic environment. In order to address this problem, this paper proposes a fingerprint indoor positioning method integrating Bluetooth and geomagnetic field (e.g., areas of the pending a set number of Bluetooth beacons). Then, Bluetooth and geomagnetic signals are respectively received through mobile phones bring iBeacon function and hall sensor, then user locations can be estimated by the fusion of Bluetooth and geomagnetic filed. According to the method, the positioning accuracy can be improved under the condition that additional other hardware resources are not added. Simulation results show that the fingerprint indoor positioning method has higher estimation accuracy.

Intelligent fault demarcation and locating technology for 5G base stations is a key technology for intelligent wireless networks. Currently, base station fault analysis relies on expert experience, board status, base station power environment data, and base station fault types, which is inefficient. The problem is not handled in a timely manner, which greatly affects customer experience. This paper uses the historical alarm data of 5G base stations and the marked data of historical processing alarm work orders, uses TF-IDF and Word2vec technology to create feature information data for alarm information, and uses Catboost algorithm to build AI model to automatically learn the hidden feature information in the data. Intelligent demarcation and locating of 5G base station fault alarms and prediction of handling solutions.

For 5G telecom operators, most of SA network adopts the coexistence of VoNR scheme and EPS fallback scheme to realize 5G voice services. Guaranteeing mobile users’ perception and experience of 5G voice services is of great significance to improve telecom operators’ core competitiveness, network reputation and user stickiness. This paper focuses on the research of mobile users 5G voice service awareness problem positioning system. Based on signaling data, the designed system periodically calculates users’ 5G voice key performance indicators, monitors indicators through specific detection algorithms. Furthermore, this system finds deterioration problems of indicators in time, and then this system locates and analyzes failure causes, and matches experience library to complete root cause positioning and intelligent production solutions. The designed system can shorten the abnormal processing time and improve users’ perception of 5G voice services.

It can bring about innovation of network and new network architecture for big data technologies, artificial intelligence technologies, network awareness technologies, network self-restructuring technologies and other technologies to be applied into network area. Introducing human-like features into network can make the network make full use of advantages of human being and bring a new network architecture named human-like networking to come into being. This paper specifies the architecture, features and mechanism of human-like networking.

Customer satisfaction prediction is a very important and complex task for the long-term development of the enterprise. Predicting customer satisfaction can provide accurate insight into customer behavior and turn post-event retention into pre-event attraction, thereby increasing brand loyalty and facilitating business conversion. This paper introduces the construction process of customer satisfaction prediction model, through the collection of B-domain, O-domain, and C-domain data, which is conducive to big data analysis to clarify the benchmark values of key driving factors. Furthermore, this paper classifies users according to four dimensions, and forms a complete preliminary model of customer satisfaction prediction. The XGBoost algorithm was introduced to iteratively upgrade the model carrying out comprehensive analysis, realizing the role of service evaluation and supervision for telecom operators. The practical application of this model is of great significance for improving the brand awareness and word-of-mouth communication of telecom operators.

With the development of Markov’s theory, Markov’s model has been widely used in different fields such as stock, electric power and environmental management because of its stationary characteristics. In this paper, the Markov prediction model is innovatively applied to predict the spatiotemporal changes of 5G hotspot users. The pattern optimization mode is used to dynamically adjust the network structure, which can effectively improve network efficiency and user experience.

With the installation and operation of large-capacity wind turbines in China, accidents such as fan collapse and blade tower sweeping have occurred from time to time, and there is an urgent need to carry out research on swing sensing detection technology for large wind turbines and towers. We have designed a real-time monitoring platform architecture for the operation status of wind turbines based on BDS (BDS Navigation Satellite System) and 5G (5th Generation Mobile Communication Technology), as well as advanced imaging LiDAR. In response to harsh application environments such as wind turbine vibration and blade obstruction, we use BDS three frequency dual antennas for precise pose measurement; and use a rigid pose transmission model to achieve pose transformation to imaging LiDAR. Then, we calculate the distance between the fan blades and the tower through the blade safety clearance area monitoring model. Finally, experimental verification was conducted on the northeast mountainous wind field. Under high-frequency vibration conditions, the horizontal positioning accuracy was 5 cm + 1 ppm, the vertical positioning accuracy was 15 mm + 1 ppm, and the average deviation of heading accuracy was 0.05°. The safe clearance area monitoring accuracy based on imaging LiDAR was better than 3 cm.

Atmospheric duct effect is an effect that occurs in the troposphere, where the atmospheric layers change with the altitude due to temperature or water vapor, forming a thin layer of the atmosphere, in which the electromagnetic waves transmitted are confined. This phenomenon is called atmospheric duct propagation of electromagnetic waves. In the slot structure of the NR TDD network, GP channels are set to isolate uplink and downlink timeslots to avoid mutual interference between uplink and downlink signals. In the atmospheric duct effect, signals from the interfering base station are transmitted over a long distance, generating remote co-channel signals, causing severe uplink interference. Subscribers cannot access the network or use the network. Based on the remote interference management method for atmospheric duct effect in the NR TDD system in 3GPP specifications, this document further analyzes the proactive identification measures and corresponding collaborative avoidance methods for atmospheric duct interference.

Accurate prediction of service capacity is very important for network optimization and network resource scheduling. It is difficult to predict the service capacity accurately because of the mobility of the population cycle and the complexity of the environment in the present network. Aiming at this problem, this paper proposes an algorithm for forecasting service capacity index based on attention mechanisms that can effectively mine the periodicity of service capacity index. Based on the STL model, the trend items and seasonal items of the capacity index series are decomposed, and then the improved Transformer attention mechanism model and the LSTM model are constructed. Collect and build the service capacity index database, build data sets for a variety of typical scenarios, and divide the training set and verification set. Based on the proposed algorithm, the uplink traffic indicators were modeled, and the average absolute percentage error (MAPE) of the model was reduced to 15.0%. The model accuracy was improved by 5.7%, and the R2 coefficient indicator was 0.903, which is significantly better than other comparative prediction models.

The power consumption of 5G base stations is a major pain point for operators, 5G energy-saving strategies are currently simplistic, it usually sets a unified energy-saving time periods, which can not achieve differentiation strategy in different scenarios. In this paper, we propose a new energy saving strategy prediction model based on XGBoost. First, the energy saving methods for 5G base stations are briefly described. Then, the energy-saving network elements are introduced to dynamically and uniformly manage the energy consumption of the whole 5G access network. In addition, the principle of XGBoost algorithms is further analyzed. Finally, the model training process and application effects are roughly proposed, which reduces the false shutdown rate to 2% and increases the recall rate to 87%. It realizes the intelligent identification of energy saving scenarios in the realistic wireless network. The intelligent recommendation of energy saving strategies achieve intelligent energy saving effects and ensure that the user experience is not affected.

This study focuses on adding channel and spatial attention modules to a residual block for more effectively interpreting high-resolution remote sensing images. Three feature extraction networks are constructed through parallel and sequential combinations to select a network for high-resolution remote sensing images. Furthermore, these three networks are combined with classification, semantic segmentation, and object detection networks. The results of ablation and comparative experiments on different remote sensing data sets—NWPU-RESISC45, DeepGlobe, and DOTA—reveal that the three constructed methods have achieved good results in semantic segmentation and classification tasks. The construction method of sequential combination (spatial–channel) is the most improved. The effectiveness and general applicability of the proposed feature extraction networks are verified. These feature extraction networks can be used in classification, semantic segmentation, and object detection tasks for remote sensing images and can effectively improve the accuracy of the network.

According to WHO statistics, fall death is the second leading cause of accidental death, among which the elderly have the highest risk of death or serious injury due to falls. Real-time detection and accurate medical treatment after a fall can effectively reduce the risk of death. At present, the implementation of detection systems for detecting human fall is mainly based on the following three categories: wearable devices, monitoring systems based on environmental sensors, and visual monitoring. Wearable devices require contact with the body, making them difficult to operate and not highly acceptable; The monitoring systems based on environmental sensors are mainly infrared radar and millimeter wave radar, and the accuracy of fall detection is currently relatively low; Visual monitoring is currently mainly based on monocular cameras, which cannot obtain depth information and has low accuracy. This system combines the advantages of machine vision and multiple sensors, and has been upgraded from a monocular camera to a binocular camera, which can more accurately obtain depth information and improve fall detection accuracy.

With the vigorous development of 5G network and the gradual withdrawal of 2G network, the current voice service research focuses on VoLTE (Voice over Long-Term Evolution), EPS FB (Evolved Packet System Fallback) and VoNR (Voice over New Radio). The coexistence of multiple voice schemes leads to complex network formats and frequent switching between networks. Therefore, the difficulty and demand of voice service network optimization have increased. Screening cells with poor voice service’s quality is one of the most effective means for network optimization. In this paper, a voice service monitoring system for 4G/5G networks is designed to monitor and screen cells with poor voice service’s quality. Through the verification of existing network data for 5 consecutive days in city X, the system can accurately screen out key cells with poor voice service’s quality, which indicates that the system is practical.

Engineering parameters play a pivotal role in the optimization of wireless networks, as they constitute the bedrock upon which successful planning, optimization, and various government or enterprise services rely. The precision of these parameters directly influences the effectiveness and efficiency of wireless network operations, elevating their significance to paramount levels. The cornerstone of wireless network optimization planning lies in the precise engineering parameter data. In this article, our primary focus revolves around the meticulous validation of two crucial parameters in wireless engineering: the antenna’s latitude and longitude coordinates, and the azimuth angle. Starting from MDT (Minimization Drive Tests), the MDT data is converted into image data, and then artificial intelligence image processing technology is used to verify the engineering parameters, thus supporting the engineering verification work and laying a good foundation for improving network quality and location service capabilities.

Nowadays, with the application of 5G-related technologies, the mobile terminal market changed the direction to develop the dual SIM phone in domestic. Thus, identifying the primary and secondary card of dual card 5G terminal users is quite important for every operator. This study proposed two algorithms which are AI brand Recognition Algorithm and Spatio-temporal Adjoint Algorithm, aim to implement dual IMEI for 5G dual cards phones. Big data analysis is used to explore the traffic threshold which will be used to define primary and secondary card users. During the research, 5G Full TAC Terminal Library, terminal data of Ministry of Industry and Information Technology and Global System for Mobile Communication Association (GSMA) are used to be the basis databases, which can identify the terminal is 5G or not. The result of study is find the traffic threshold and then define the card type of China Unicom 5G users. As the result shows, China Unicom had 5G network users 150.1 million totally, in which single SIM users accounted for 0.58% of the total number of users, and dual SIMs users accounts for 99.42% of the total. The above research will reference for the primary and secondary card marketing.

5G core network realizes three-layer decoupling based on virtualization technology, and the architecture is more flexible and complex. However, once various interfaces and network elements generate alarms, alarm correlation and root cause analysis are difficult, which brings challenges to operation and maintenance. This paper proposes a method of alarm correlation and root cause analysis. We establish the three-layer correlation relationship of resource and alarm data, and set up the alarm correlation rule base based on expert experience and graph neural network algorithm. In the method, the alarm data can be compressed, and the resource topology and alarm root cause can be viewed. The method can greatly improve operation and maintenance efficiency.

With the 5G network’s development, complaints about the 5G network experience are increasing. 5G uplink and downlink capability analysis, uplink and downlink balance data analysis, and terminal capability analysis on the co-construction and sharing platform based on big data and blockchain, it is determined that “uplink coverage limitation is earlier than downlink coverage limitation” is the main factor affecting 5G user experience. Through the in-depth analysis of the network, three methods such as uplink enhancement, the problem of uplink limitation is solved and the user perception is improved.

RedCap (Reduced Capability) is a 5G technology defined by the 3GPP standardization organization and belongs to the new technology standard NR light (New Radio lite). In June 2022, the 3GPP Release 17 standard has been frozen. RedCap is also one of the important features of the R17 standard. Based on the commercial network of Shanghai Unicom, this paper studies the access of multiple BWPs and multiple BWP users in Redcap, and recommends the typical network configuration.

With the development of 4/5G network co-deployment and continuous updates of mobile applications, wireless network users’ usage habits are constantly changing, and the quality requirements for network services provided by operators have become more stringent. Therefore, there is an urgent need to shift our evaluation of network quality from being based on network itself to focusing on customer perception and quality assurance mechanisms. This article proposes a method and tool for evaluating the performance of wireless networks based on multi-source data, primarily wireless trace data and Over-The-Top (OTT) data. The aim is to introduce innovative multi-source data, which has not been thoroughly researched in the current stage, to build a new network evaluation system on the existing foundation. This system aims to make the evaluation results more closely aligned with users’ real network usage experience, enabling more accurate guidance for operators’ network optimization work and target areas of improvement.

In the face of the national “double carbon” strategic goal, major operators, major equipment manufacturers and professional vendors have developed various energy saving algorithm models and application services, and the quality and availability of the underlying data is the key to the success of these models and services in the overall energy saving system. Given the characteristics of a variety of data involved in the energy saving system, such as cross-discipline and heterogeneity, this paper proposes a unified data collection framework and system that meets the requirements of large-scale data collection such as high reliability, high availability and high concurrency, while supporting on-demand access of third-party collection adapters with strong scalability. The pilot application results of this system show that this system can make better use of server hardware resources and significantly reduce the end-to-end latency of data collection in the high-volume data collection scenario.

To estimate the precipitation amount from communication signal attenuation is a relatively novel and promising method for its good measurement accuracy and high spatial-temporal resolution. In this paper, we will introduce a 5G MR information precipitation estimation algorithm to predict the precipitation. The algorithm comprises two stages: precipitation database construction and precipitation estimation. During the first stage, the precipitation database undergoes pretreatment through limiting and moving average filtering. In the subsequent online matching stage, the rain intensity is calculated using the support vector machine (SVM) algorithm. The results show that the average accuracy rate has reached 94.84%.

As an important transportation hub, the airport plays an important role in emergency prevention and control; In the process of digital transformation of enterprises, technologies based on big data architecture help airports realize digital transformation. Airport big data screening system for special passengers Design and implementation based on SQL server 2008. Based on passenger big data, blockchain, and virtualization of cloud platform technology, it can realize dynamic analysis and prediction in advance, help the airport group to respond quickly to emergency prevention and control, and achieve remarkable results in the transformation of scientific and technological achievements.

With the widespread adoption of 5G networks, more and more network operator equipment now supports 8 transmit and 8 receive (8T8R) antennas on the base station side. However, the cost of remote radio unit (RRU) equipment of 8T8R is high. To address this issue, this paper proposes a method to revitalize old 4T4R equipment and reuse existing 5G network sites by combining two 4T4R RRUs. Additionally, to solve the problem of weak coverage in low- and middle-rise residential areas, the paper introduces an RRU channel soft correction algorithm, which can correct RRU channels without additional equipment, and verifies its effectiveness in a real network. The results show that this scheme can effectively improve indoor depth coverage capability and customer network experience.

To address the problem of data confusion and regulatory difficulties in network freight transportation, a network freight bill abnormal detection method based on a Belief Rule Base (BRB) and Evidence Reasoning (ER) algorithm, and incorporating Differential Evolution Algorithm (DEA), is proposed. Firstly, to reduce the complexity of the BRB model, relevant features are selected for input into the model, and a BRB abnormal detection recognition model is established. Secondly, ER inference rules are used to reason the belief rules, and DEA is used to optimize and adjust the parameters of the BRB model. Finally, the experimental results are compared with artificial data and other non-BRB methods such as neural networks and support vector machines. The results show that the proposed abnormal detection model can effectively identify abnormal data, and some of the rules have been applied to the Network Freight Transportation Information Exchange System of the Ministry of Transport.

The uRLLC scenario is one of the three major 5G application scenarios and will become an important scenario for 5G network evolution in the next few years. The uRLLC scenario supports ultra-low latency and highly reliable communication services. As the uRLLC scenario becomes more mature, operators need to analyze service latency from an end-to-end perspective. Firstly, this paper analyzes the composition of end-to-end delay of 5G network, and points out the direction of 5G network delay voltage drop based on the characteristics of high ratio and large jitter of air interface delay. Then, the wireless side technology schemes suitable for uRLLC scenario are introduced, and the effect of the schemes is analyzed. Next, the typical service delay and reliability requirements of uRLLC scenario are introduced, and the deployment suggestions of each technical solution of typical services are further put forward. Finally, through specific practical cases, the technical basis for expanding 5G convergence applications in vertical industries is provided.

With the rapid development of 5G network, network architecture is becoming increasingly complex, and the number of network devices are also increasing significantly. Complex and diverse network data is generated during the daily operation of network devices. In addition, network alarms seriously threaten the network operation, and network outliers clearly indicate the existence of hidden dangers in the network. In order to guarantee effective network operation, this paper investigates two algorithms about alarm cleaning and outlier detection. An alarm cleaning algorithm is proposed and based on network function association relationship, which can effectively filter out the redundant alarm and facilitate intelligent alarm analysis. In addition, an outlier detection algorithm is proposed and based on the standard interquartile range (IQR) method and the outlier coefficient of data temporal correlation, which is helpful in detecting network outliers more effectively and facilitate network performance analysis and automatic operation and maintenance.

The issue of 5G energy efficiency is not only about energy saving, but also about business experience. The yardstick for measuring green networks has changed from “energy consumption” to a comprehensive test of “energy efficiency”. How to construct a scientific and reasonable network energy efficiency evaluation method, is an urgent problem to be solved. In this paper, a reliability index is proposed to characterize the network performance of URLLC service, and the energy efficiency of URLLC network is calculated based on the reliability to improve the evaluation criteria of energy efficiency.

As a typical scenario of 5G network, eMBB is designed for public UEs to provide broad-bandwidth and high-bitrate communication service. UE’s throughput is the most important performance indicator of a network providing eMBB service. This paper investigates the integrated algorithm of UE perception throughput from the management aspect. The algorithm takes all the transmission data including tail packets and small packets into consideration based on equivalent normalization to better reflect the real bitrate UE experienced.

NSA (Non-Standalone) network plays a key role in 5G construction, which can accelerate the popularization of 5G and enable us to use 5G faster and with less investment. But the unique networking mode of NSA also brings some difficulties to the evaluation of wireless network quality. Currently, in the NSA networking mode with dual connectivity, the MR (Measurement Report) of LTE (Long Term Evolution) MCG (Master Cell group) and NR (New Radio) SCG (Secondary Cell group) are reported separately. There is no relevant information and data of the other party in the MR of the MN (Master Node) and SN (Secondary Node), and correlation analysis cannot be conducted. The NSA wireless network quality evaluation method proposed in this article can effectively solve this problem. Firstly, from the associated MR of NSA MN and SN by tracking the SN addition procedure in the trace data. Then, evaluate the necessity of MN configuration and the delineation of NR perception issues based on the statistical analysis of associated MR.

In order to achieve the goal of “Carbon Neutral” and “Carbon Neutral” in the long-term coexistence of a multi-mode communication network and its high energy consumption, it need to put forward corresponding measures in energy saving, emission reduction and resource recycling, to realize high-quality development of mobile network, energy saving, cost reduction. The integration development scheme of mobile networks and fixed networks using small stations can make use of existing resources, maximize the network efficiency of operators and enhance user perception.

Digital transformation has become an important engine for improving the quality and efficiency of all walks of life. As an important part of the digital economy, identifying and making full use of data elements is the key to promoting data value transformation. This article is based on the data value system and characteristics of operators and based on the cost value theory. It comprehensively considers the economic value of data, its utility to users, and market regulation, and proposes a data asset pricing model suitable for operators.

With the development of information technology and the growing communication needs of users, telecom operators have launched various products for the individual market. How to recommend the appropriate products to the right people in a timely and accurate manner is the key to increasing revenue and improving user experience. The previous works mainly based on product features to subjectively filter target users or create traditional machine learning models by integrating different features, the former is limited by subjective influence and has low accuracy rate, the latter is incapable of digging deeper into the hidden meaning of feature crossover, so none of these methods can accurately capture the marketing opportunity brought by user behavior changes. In this paper, we obtain user portrait data and user behavior data, and use Transformer model to capture the deeper intent in user behavior sequences in order to provide accurate and timely services to users, and finally demonstrate through experimental results that the model outperforms traditional Random Forest and LightGBM algorithms.

In the 5G era, network slicing offers customized and dedicated network services, which enable independent and isolated operations for different businesses, making it a pivotal aspect for the diverse range of services. During the mobility of slice users, cell handovers are unavoidable. However, relying solely on channel quality and the number of slices in the target cell as criteria could potentially result in users being switched to target cells that cannot support the ongoing critical slice services, and this will significantly impact user perception. To address above mentioned challenges, this paper proposes a novel handover method based on the priority of slice services. The proposed method can guarantee that the perception of important and necessary slicing services is prioritized. Meanwhile, the proposed method can also maximize the user perception of handover in a scenario of concurrent multiple slices. By employing this handover method, service perception and user experience can be effectively enhanced during slice user mobility.

In order to study how massive data with periodic and tidal performance influences identification and processing in present network, this paper proposes a data-processing method combined with deep learning algorithm, XGBoost algorithm and spatio-temporal network model. By collecting a large amount of data from existing network and dealing it with pre-treatment, such as anomaly detection and periodicity detection, this proposed data processing method is capable of weeding out erroneous data and fulfill whole data set. In this way, we are capable of getting some characterized results and forming a result set. XGBoost algorithm and space-temporal network model, can then help with data analysis and providing reference for subsequent research such as feasibility analysis and scheme design. This scheme proposed in this paper is able to work in an effective way, especially when facing problems composed of large amount of data with periodicity.

This thesis mainly focuses on the process of upward aggregation of each layer in the construction of O-domain labeling system and the rationality evaluation system research and system development of several points in the design of basic forms, data intermediate forms, data indicators, and data labeling sharing patterns [1], specifically including metadata unification, aggregation of each layer, data heat, dimensional redundancy, cross-tenancy, and inter-platform unification. The research ideas of this thesis are mainly developed from two aspects: metadata acquisition and buried data acquisition, and data heat analysis.

With the sustainable development of transportation, the network construction along transport lines has received increasing attention. The network coverage effect of base stations along transport lines is a significant issue. This paper proposes an evaluation method based on transport lines layers and the PostGIS spatial database. Firstly, a method for modified transport lines layers is proposed. Secondly, on the basis of transport lines layers and distribution of surrounding base stations, covered transport lines layers can be obtained by geographical computation. The original length and covered length of transport lines can be calculated and the coverage is given. Areas with weak coverage can be accurately located through geographical display, thereby providing precise data support for network planning.

Full digital aperture has the advantages of flexible beam number, precise beam pointing angle, and strong anti-interference ability. It outperforms analog phased array aperture in terms of simultaneous multiple transmitting beams. The efficient measurement of receiving and transmitting multi-beam patterns for full digital apertures is difficult in the field of digital antenna measurement. Near-field measurement for radiation patterns is an important test method in the field of antenna measurement. This paper proposes an automatic measurement method for simultaneous digital multi-beam patterns based on the existing near-field scanning framework. The digital aperture generates a test channel by itself, and constructs a reference channel that is fully coherent with the test channel, extracts the amplitude difference and phase difference at multiple frequency signals between the test channel and the reference channel, and generates corresponding frequency signals in the scanning coordinate system. Amplitude and phase distributions are sampled and transformed to obtain the corresponding frequency signals and radiation pattern, avoiding the need for additional equipment to achieve multi-frequency point transmit beam measurement at the same time. Finally, the measurement of a 128 digital channels aperture agrees with simulation, which verifies the effectiveness of this method, and provides a new quality capability for full digital aperture multi transmit beams measurement.