Abstract：Integration of satellite and terrestrial networks would bring better coverage, more flexible networking and more robust network architecture for 5G, which makes the 5G network connecting everything possible, and then becomes an important evolution direction for 5G network. This paper firstly introduces the necessity to support integration of satellite and terrestrial networks in B5G era, then analyses the developing phases and technical status of 5G network integrating with satellite from network architecture and technical standard point of view. Based on the technical status, the paper further analyses issues on system architecture evolution that 5G network integrating with satellite faces, and puts forward some architectural enhancement solutions, including 5G system enhancement on supporting satellite backhaul and User Plane Function(UPF) on-board.

Abstract：Mobile edge computing(MEC) is proposed to meet the growing computing demands.Driven by ubiquitous connectivity and global regional coverage, Low Earth Orbit(LEO) can break through geographical restrictions and achieve global wireless coverage, which is an indispensable choice for future 6G mobile communication systems. Therefore, this paper considers the formation of edge computing satellites by placing edge servers on satellites, and proposes a system architecture for satellite edge computing, which can provide computing services for users in remote areas on the earth or users who need emergency communication in extreme environments.

Abstract：Solar array is the only energy source for the majority of satellites, in view of the great need of the 6G space communication, under the premise that the output power is constant, satellite solar arrays need to further reduce the mass and volume. Roll-out solar array has the characteristics of lightweight, high-efficiency, and no need of the unfold drive, which has become an important direction for the development of space solar arrays. This paper introduces the development process of roll-out solar arrays and the development status of roll-out solar arrays in the domestic and foreign. The key technologies are sorted out to provide a reference for the selection of 6G satellite solar arrays and the development of roll-out solar arrays in the future.

Abstract：The construction of digital village is a strategic direction for rural revitalization, and more importantly, an important support for the landing of digital China. The development of China′s village has characteristic political, economic and cultural forms, and also carries a unique development mission in the national strategy. Around the urgent demand of expanding domestic demand, accelerating factor flow and smoothing the large domestic circulation, it is urgent to promote the upgrading of digital village construction. The digital city has formed a good foundation and advantages in the development environment, core technology, market space and supervision guarantee, which becomes the core leading and important guarantee for the development of digital village.

Abstract：In order to promote the research of image style transfer technology based on neural network, the main methods and representative work of image style transfer based on neural network are summarized and discussed in this paper. This paper reviews the traditional style transfer algorithms, introduces the basic principles and methods of the main image style transfer based on neural network in detail, analyzes the application prospect of the related field of image style transfer based on neural network, and finally summarizes the existing problems and future research direction of the image style transfer based on neural network.

Abstract：The rapid increase of the number of cancer patients has attracted worldwide attention. Researchers are very concerned about the assessment of the carcinogenicity of compounds, but this is extremely challenging. In this paper, 341 kinds of experimental data were obtained, and the spatial atom feature combined with the spatial graph convolutional network(SGCN) was used to establish a model that could predict the carcinogenicity of compounds. The results showed that when compared to other models, the classification model of the SGCN was more suited to predicting the carcinogenicity of compounds and had an overall classification accuracy of 96.9%, which showed that the SGCN model could accurately classify chemicals and had considerable potential in practical applications.

Abstract：The existing air quality prediction methods rarely consider seasonal factors, and the prediction effect is not good. Therefore, an air quality prediction method based on improved binary chaotic crow search algorithm(BCCSA) and deep long short term memory neural network(LSTM) is proposed. Firstly, the method of seasonal adjustment is proposed to preprocess the collected original air quality data in order to eliminate the influence of season on prediction. Then, an improved BCCSA is proposed to optimize the air quality data. Finally, the self-attention mechanism is added to the deep LSTM to predict the air quality data. The experimental results show that this method can effectively improve the prediction accuracy of air quality.

Abstract：A feature screening method based on alpha wave and principal component analysis was proposed to solve the problem that the weakly correlated feature quantity would affect the classification accuracy in EEG motor imagery classification. Based on brain computer interface system, the EEG signals corresponding to left and right motor imagination tasks were generated by auditory stimulation and processed by wavelet packet decomposition, and then the α band signals of the EEG were reconstructed, so as to extract the α waveforms and extract the statistical features. Combined with PCA technology and SVM method, the weak correlation features are eliminated and classified. According to the selected data, the accuracy of the results is higher, and the accuracy of signal classification is improved from 90.1% to 94.0%.

Abstract：As the complexity of the chip increases, the time it takes to verify the chip in the design process continues to increase. Aiming at the poor reusability and low coverage of traditional verification platforms, this paper uses Universal Verification Methodology(UVM) to design the Wishbone-SPI verification platform, uses UVM components to flexibly build the verification platform, completes the standard verification framework, and designs constrained random stimulus, automatically counting function coverage. The simulation results show that the functional coverage of the verification platform reaches 100%, and shows that the platform has good configurability and reusability.

Abstract：In the finish stage of the chip physical design, in order to meet the metal density design rules, dummy metal fill needs to be added. The dummy metal fill generates external parasitic capacitances, which will deteriorate chip timing results. In process nodes above 40 nm, timing deviation caused by these external parasitic capacitances is about 0.12%, even smaller than the mismatch between STA and SPICE simulation, we usually ignore it. However, this timing deviation must be repaired in advanced process nodes with FinFET structure. Taking a industrial DSP process chip with FinFET structure as an example, this paper uses QRC to compare the parasitic capacitance changes before and after adding dummy metal fill; uses Tempus to analyze the reasons for the timing deviation of the chip; finally proposes a method for repairing timing deviation based on Innovus, the timing result is verified by signoff. This method effectively improves the efficiency of timing closure.

Abstract：The radio monitoring system puts forward higher requirements for signal search, interception and identification capabilities. The traditional radio monitoring system software and hardware have immature standardization and modularization, strong customization, and poor versatility, and it is difficult to meet future equipment needs. An open radio monitoring system based on VPX combined with the idea of software radio is designed, and the software and hardware architecture of the platform is intensively studied. Finally, a verification platform is built for testing the transmission and system performance, which verifies the feasibility of this platform.

Abstract：In order to solve the problem that the center position of the belt layer is offset during transmission, a 3D sensor based belt layer deviation correction algorithm is proposed in this paper. Firstly, the contour data collected by 3D sensor is preprocessed to make the data smoother. Secondly, in the process of correcting deviation, the belt layer is divided into three regions: upper, middle and lower regions, the upper and lower regions adopt fixed edge, and the middle region adopts fixed center. The real-time correction is carried out by calculating different positions in three different regions. Finally, through MATLAB simulation, the correction accuracy of the algorithm can reach within 1 mm, which is better than the traditional unilateral correction accuracy and correction stability.

Abstract：Aiming at the problem of estimation of long and short code direct spread code division multiple access signal with residual frequency offset, firstly, the signal matrix is constructed by using the periodic piecewise received signals of short code and long code. Secondly, the eigenvalue decomposition method is used to obtain the user eigenvector sub-space with residual carrier, and the digital phase-locked loop is used to track and eliminate the residual frequency offset. Then, the Givens matrix is used to rotate the load matrix composed of eigenvectors, and a whale optimization algorithm combined with random difference and simulated annealing is proposed to find the best rotation angle to minimize variance, so as to estimate the composite code segments of each user and splice them in turn. Finally, Berlekamp-Messay algorithm and correlation operation are used to estimate the long code and short code of each user. Simulation results verify the effectiveness of the proposed algorithm for pseudo code estimation of LSC-DS-CDMA signals with residual frequency offset.

Abstract：With the deployment and commercialization of 5G networks, providing differentiated services for different users in the existing network has become a research hotspot in the industry based on the different needs of the business for network capabilities. Therefore, this paper proposes a method to realize the converged deployment and unified invocation of 4/5G network quality of service(QoS) differentiated services. This scheme includes the network architecture of 4/5G network QoS differentiated service fusion deployment, the method of QoS platform server to terminal 4/5G user automatic identification, and the end-to-end process of 4/5G converged QoS service. The design of this scheme enables the third party to directly call the QoS capability of the operator using the unified scheduling interface of the whole country without distinguishing the 4/5G form of the user, giving full play to the network advantages and value of the operator.

Abstract：In view of the safety problem of rock burst in coal production and mining, an assessment and analysis of coal mining impact model based on hidden Markov(HMM) model was proposed. Data samples collected under normal and abnormal conditions in coal mining environment were selected as offline training samples to obtain parameters of pre-mining environment model. Then through real-time detection of coal mining data samples as online training samples, real-time data is sent to the pre-mining environment model for evaluation, and the maximum probability is obtained according to the maximum likelihood learning, and the current operating state of the system is obtained through matching. The experimental results show that the model has high reliability in the coal mining impact and the coal safety monitoring work is stable.

Abstract：Three dimensional point clouds are widely used in robots and automatic driving. The research results of deep learning on two-dimensional images are remarkable, but how to use deep learning to identify irregular three-dimensional point clouds is still an open problem. At present, due to the complexity of the data of the scenic spot cloud itself, the uneven distribution of points caused by the change of the scanning distance of the point cloud, and the challenges caused by noise and abnormal points still exist. Aiming at the problems of low classification efficiency and low classification accuracy of the existing deep learning Network framework for laser point cloud data, a CNN Transform framework based on laser point cloud feature image and Light-BotNet is proposed. The framework is to extract the features of point cloud data, construct the point cloud feature image with adjacent feature points as the input of the network framework, and finally take Light-BotNet as the network framework model for point cloud classification training. The experimental results show that compared with most existing point cloud classification methods, this method can better improve the classification efficiency and accuracy of laser point cloud.

Abstract：In order to make the response of the temperature control system more stable, the third-order discrete filter is introduced in PID control algorithm to filter the input signal, which eliminates the jitter noise and makes the outputs more steady. Besides, a temperature control system based on STM32 micro-controller is designed. The temperature acquisition module is composed of platinum resistance Pt1000 and MAX31865, and the temperature regulation module includes semiconductor cooler TEC and H-bridge logic. The computer software communicates with the cotrol system through USB serial port. The received real-time temperature data is used in PID control algorithm by introducing filter, and the output value is afterwards sent to the computer software to realize a closed-loop temperature control process. The experimental results show that the PID algorithm with filter outperforms PID combined with Bang Bang, differential first PID, step PID and variable integral PID, which improves the ability of anti-interference and has the practical value and application value in engineering temperature control.

Abstract：Information security is an increasingly concerned issue. As the carrier of information transmission, digital image contains a large amount of information. In order to improve the security of digital image transmission, this paper improves the Clifford system, analyzes the chaotic characteristics of Clifford system through chaotic attractor graph and Lyapunov exponent, and designs a new image encryption algorithm based on the improved Clifford system. Firstly, the plaintext is transformed into a binary matrix. The rows and columns of the binary matrix are circularly shifted. Then, the plaintext is divided into nine matrix blocks with different sizes, and each matrix block is scrambled. In block scrambling, chaotic sequences are used to scramble between blocks, while cyclic displacement scrambling is used to scramble within blocks which ensures that the position of each pixel has changed. Finally, the image is diffused to get the ciphertext. Simulation experiments and security analysis show that the algorithm is safe and reliable for image encryption.

Abstract：A fear emotion inducing system is designed and implemented based on high-altitude VR interactive scene, which can provide lower limb somatosensory interaction. By the customized carbon structural steel springboard and its double barrel spring unit at the bottom, the system can produce human multi angle stress shaking after walking and trampling, and also can realize the shaking interaction in the virtual high-altitude scene with matched VR tracker. The system has the function modules of interactive display, pedal matching, virtual scene and experiment management, which can automatically store the physiological data of human, spatial position，time information and task performance during the induction process.

Abstract：With the complexity of quality inspection information management business，the existing quality inspection information management systems have so many problems, such as insufficient processing capacity for big data, high manual operation error ratio and poor scalability. This paper proposed a quality inspection information management system based on Cloud Native, then analyzed and researched on the Cloud Native technology, and clarified the principles and characteristics of Cloud Native technology. Combined with the actual use of quality inspection information management business, the Cloud-Native-based quality inspection information management system was designed and implemented in detail. Finally, the function and performance of the system were tested to verify the effectiveness of the overall design scheme.

Abstract：Because LoongnixOS is the time-sharing operating system, it cannot meet the requirement of high reliability in industrial automation application which requires to use real-time operating system(RTOS). On the basis of analysis of and study on defects of real-time performance in Linux kernel of LoongnixOS and the method to optimize real-time performance in RT-Preemption patch, the method of real-time transformation of LoongnixOS and optimizing real-time performance of LoongnixOS is put forward, in which a RT-Preempt-Linux kernel is a substitute for original Linux kernel of LoongnixOS. Self-developing real-time performance softwares and Cyclictest are utilized to verify the method. Results of test show that the method can be effective in improvement of real-time performance of LoongnixOS. Real-time performance, including process-switching time, thread-switching time and task-switching delay, of the LoongnixOS transferred to RTOS achieves the grade of microsecond.

Abstract：Functional electrical stimulation has become a commonly used and effective method for the rehabilitation of stroke in hospitals. However, due to the large size of the therapeutic apparatus, the fixed training location, and the single electrical stimulation mode, a wearable four-channel functional electrical stimulator based on the STM32 chip was designed to solve the above problems. It needs to be able to control the electrical stimulation mode and parameters in real-time, and also to meet the upper limb rehabilitation training of patients at home. The electric stimulator system includes the STM32 main control chip, power management module, boost module, electric stimulation drive circuit, and serial screen human-computer interaction module. Experimental results show that the instrument can effectively cause muscle contraction, output electrical stimulation waveforms correctly, and achieve the curative effect of rehabilitation therapy.

Abstract：The development of digital technology is changing rapidly. In the new generation of server products, the PCIe Gen5 signal rate has reached 32 Gb/s. In order to ensure its signal integrity design, single-board loss monitoring is essential, and a good Coupon design is therefore particularly important. Traditional Coupons mostly use Litek probe technology, and wiring in a broken line. Although the technology is mature, there are still many problems such as large space occupation and poor data stability. Based on the low-cost requirements of product design, this paper researches on the low-cost Coupon design of PCIe Gen5 signal. Different wiring methods are analyzed and compared through theoretical analysis and simulation. Then the high-reliability Coupon wiring method is selected which can also takes up less space. The test fixture is also optimized which can also improve data stability. Experimental results show that the Coupon design method proposed in this paper not only takes up less space than traditional methods, but also has higher data stability. Therefore, this method is more conducive to the low-cost and high-reliability design of Coupons.

Abstract：In order to improve living conditions in the town and solve the problems from garbage floaters on small-area water surface in landscape zones, this article, based on Faraday waves theory, designs a small-area water floaters collection automatic system device, which is consist of vibration source(to produce Faraday waves), floating platform and collecting device. This device utilizes single chip to output specific frequencies to the floating platform, and then makes the floaters which connected the floating platform to clap the water surface in order to produce waves in corresponding wavelength. This device is able to grab long-distance granules floaters on water surface. Under Faraday waves′ characteristic effect, granules floaters on water surface will move to a certain place according to relevant track, as a result, it is convenient for collecting device to salvage and recycle water surface garbage. This research result is able to collect granules floaters on small-area water surface without any second harm to the organism in the water. It is an economic, environment-friendly innovation.

Abstract：In the field of traffic Internet of Things, there is a business demand for road traffic flow information monitoring. The ground sensing coil is arranged on the road to sense the information of vehicles entering and provide initial data for the traffic Internet of Things. According to the working scene of ground induction coil, the working principle of ground induction coil is described. On this basis, a circuit of vehicle entry detection signal based on ground induction coil is designed. The signal acquisition circuit is mainly composed of tuning oscillation circuit, waveform shaping circuit and microprocessor. The design points of each module are introduced. PSpice software simulation shows that the circuit can effectively collect the state value change of ground sensing coil, so as to detect vehicle entry. It has high practical value and application prospect in the field of traffic Internet of Things.

Abstract：The Spherical Centroid Voronoi tessellations(SCVT) are a kind of high-quality unstructured meshes, which are adopted by well-known earth system models at home and abroad. Internationally, the open source project Message Passing Interface SCVT(MPI-SCVT) has realized the parallel SCVT generation. However, although MPI-SCVT works well for the generation of general-resolution meshes, the use of MPI-SCVT presents several problems when the required mesh is of very high resolution, including the poor quality of initial mesh points, large data MPI communication problems, external library calling problems and I/O problems. For these problems, a set of solutions are proposed. Practice shows that the application of these solutions can successfully complete the generation of the global 1.9 km resolution grid, and the number of iterations is significantly reduced when reaching the same accuracy, which saves the calculation time.