Abstract：Molecular dynamics (MD) simulation is an important tool to explore the microcosmic world, which is widely used in many fields. Two-dimensional materials is an important research direction of MD in the field of materials science, in which the calculation of interlayer interactions is the most time-consuming part. High performance computing is a key technology to improve the simulation efficiency of two-dimensional materials. In this work, the powerful computing power of a new generation of Sunway supercomputer is utilized to improve the MD simulation efficiency of two-dimensional materials. For the interlayer force field, multiple algorithm optimization strategies such as eliminating redundant calculation, multi-core fusion and setting buffer are adopted. It implements thread-level parallelism by software Cache accumulation force, communication between computation processing elements accumulation energy and C++ feature. The hardware and software Cache coordination policy is adopted to improve storage access efficiency. The experimental results show that the overall performance is improved by 155 times, the simulation efficiency is about 2 ns/day, and the weak extension parallel efficiency reaches 92.7%.

Abstract：The autonomous driving simulation platform is an effective means to solve the problems of long actual vehicle testing time, high cost, and difficulty in reproducing extreme scenarios in autonomous driving. With the introduction of large-scale cloud computing and high-performance heterogeneous computing, the simulation platform faces the challenges of efficient virtualization, balanced scheduling, and convenient end cloud interaction. Therefore, this article designs a parallel system architecture that supports lightweight virtualization, and integrates fine-grained resource balancing scheduling and low latency remote interaction methods. It constructs an ADsim high-performance parallel simulation platform, which has been tested and verified to have high performance and convenient interaction characteristics.

Abstract：With the deepening of neural network layers, the sparse deep neural network has more advantages in computing and storage space, but the performance of the sparse deep neural network still needs to be optimized. Therefore, a performance optimization method based on GPU sparse deep neural network is proposed, which adjusts the order of computation, enhances the reusability of data, and combines the unique structure of GPU with CUDA programming method, performance is further improved by prefetching and other methods. According to GraphChallenge's official data set, it achieved up to 2.5 times the performance acceleration compared to the related cuSPARSE library functions.

Abstract：This paper proposes a general CNN hardware accelerator design scheme based on FPGA. For the most computationally intensive convolutional layer, three acceleration modes are adopted: input channel parallelism, intra-core parallelism, and output channel parallelism, and the corresponding parallelism degree is reasonably set according to the on-chip resources of FPGA. In terms of data loading, adjacent data bit width combined transmission is adopted, which effectively improves the actual transmission bandwidth of the accelerator. Based on the idea of row-based data flow loading, the input cache module is designed. The cache module only needs to cache two rows of data to start the convolution operation, effectively advancing the start time of the convolution operation. Between the data input, data operation, and data output modules, the pipeline cycle optimization method is used to greatly improve the computing performance of the hardware. Finally, the accelerator is applied to VGG16 and Darknet-19 networks, and experiments show that the computing performance reaches 34.30 GOPS and 33.68 GOPS, respectively, and the DSP computing efficiency is as high as 79.45% and 78.01%.

Abstract：In order to improve accuracy, image semantic segmentation networks often use complex convolutional layers as the basic feature extraction units. The different types of convolutions present in these convolutional layers increase the difficulty of parallel acceleration computation for the network. A parallel computing accelerator based on FPGA for multi type convolutions is proposed to meet the accelerated computing requirements of different types of convolutions in semantic segmentation networks. Firstly, the calculation principle of convolution is analyzed. Then, based on the basic operation principles of different convolution types, a processing unit for multi multiplication parallel computing is constructed. The convolution is accelerated through multi processing unit parallelism, data reuse, and PIPELINE method. The experimental results show that for specific size feature maps, using the proposed convolutional accelerator design method can achieve a maximum speed increase of 113 times.

Abstract：Aiming at the problems existing in the current coal mine monitoring data acquisition system, such as inconsistent data coding standards, poor single system operation reliability, data visualization tools that do not support the Web mode and the host cannot be switched intelligently, based on the industrial internet platform technology system, the front and rear end separation and Restful API interface access technologies are used to integrate Internet open source reports and other components, Research and design a microservices oriented coal mine production monitoring data acquisition system, introduce the existing problems of the current coal mine monitoring data acquisition system, complete the architecture design of the microservices oriented coal mine monitoring data acquisition system, and specifically elaborate the key technologies such as data unified standardization, single channel acquisition microservices processing, data service status monitoring and data and information visualization components. The coal mine production monitoring data acquisition system has been verified in Shanxi Tiandi Wangpo Coal Mine. The application results show that the data model is designed according to the unified standard specifications, and the coal mine production monitoring data acquisition and monitoring are realized through the deployment and operation of data acquisition, data service, visualization and other application microservices, which provides important data support and basic support for other applications in the coal mine. At the same time, the operation mode of microservices improves the operation and maintenance efficiency of coal mine system maintenance personnel.

Abstract： Beam tracking is one of key technologies of millimeter wave communication systems for maintaining a stable radio link. Based on the downlink beam management procedure P1,P2,P3 and uplink beam management procedure U1,U2,U3 specified in 3GPP (3rd Generation Partnership Project) protocols, this paper analyzes and compares the resource occupation rate and beam failure probability of different beam tracking schemes. Analysis and simulation results show that the P2+P3 scheme is optimal in the sense of average resource occupation rate as the downlink resource is usually abundant in general, whereas, in the sense of beam failure probability, the U2+P3 scheme is optimal since it is free of power information feedback. In practice, an appropriate beam tracking scheme can be chosen according to the special communication requirements.

Abstract：The ZUC-256 algorithm architecture is analyzed，and a couple of implementation architectures suitable for FPGA are proposed for the multiple-input modulo 231-1 adder. By analyzing the critical path, various optimization schemes such as full hardware and hardware-software codesign are proposed, which can be applied to different application scenarios. Taking Zynq-7000 FPGA for example, the performance differences of each scheme are compared in detail. It has been verified that the proposed scheme can be applied to both ZUC-128 and ZUC-256 algorithms, and can be widely used in 4G LTE and 5G-NR mobile communication.

Abstract：Integrated sensing and communication (ISAC) is one of the key research directions of the 6G network. At present, the industry focuses on the implementation of the sensing of base stations and terminals. How to enhance the architecture of the 6G core network to support the sensing function, how to distribute the sensing function in the 6G core network, and how to open and release the sensing result to third parties are still in the early research stage. This paper proposes a key technical implementation solution of the ISAC in the 6G core network, including the distribution of the sensing functions in the 6G core network. The 5G service based architecture(SBA) is enhanced, and a dual-bus solution is added. The original sensing data is transmitted by using a data channel interface (DCI) . Sensing measurement data or sensing results are transmitted through the service-based interface (SBI). And the implementation methods of the end-to-end sensing flow and exposure are proposed.

Abstract： Compared with 4G networks, 5G networks have significant advantages in bandwidth, delay, security, reliability and other network performance. With the popularization of 5G industry application, the enhancement of 5G network capability is becoming more and more important for network service capacity expansion and business differentiation guarantee. In particular, 5G LAN research initiated by 3GPP has helped promote the combination of 5G network and industry application. This paper has in-depth analysis of 5G LAN technology solutions for 5G 2B business. It includes comparison of application scenarios, key technologies and application schemes in 5G 2B business. Finally, implementation suggestions are given based on differentiated industrial application requirements.

Abstract：At this stage, there is a lack of active planning of communication base station infrastructure, and there are multiple sets of superimposed facilities. The power supply, backup power and space are not matched, the operation lacks refined management, the ineffective energy consumption is high, and the system operation efficiency is low.O&M has not yet achieved intelligent O&M, which is inefficient and costly, and lacks the management and evaluation of the whole life cycle of equipment. Facing the vision and goal of carbon peak and carbon neutrality, this paper studies the whole scenario and life cycle logical integrated planning method for the overall planning of the "low-carbon" target network, and builds a low-carbon planning method for the whole scenario and life cycle around the base station and computer room, and continuously evolves to the goal of green and low-carbon infrastructure, so as to provide strong support for the continuous reduction of comprehensive energy consumption and the comprehensive green and low-carbon transformation and development of the economy and society.

Abstract：At present, 5G has widely used in industrial Internet, but not formed a large-scale application in the industrial field.For the quality assurance requirements of 5G in industrial applications, such as performance, functionality, service, analyzing the application requirements of 5G in different typical industrial scenarios, a quality indicator system for 5G industrial applications is constructed in three dimensions: physical parameters, network capabilities, and business requirements. Industrial users, 5G operators, and integrators can be reached a consensus on quality indicators in industrial applications. The quality indicator system is important to guide the construction and application of 5G in the industrial field, and promote the depth development of 5G industrial applications.

Abstract：Ultra-reliable and low-latency communications (URLLC) is one of the three application scenarios of 5G. It has the characteristics of high reliability and low delay. Modeling and evaluation for this scenario is significant for the design, planning and service application of communication systems. This paper focuses on the reliability evaluation for URLLC. By evaluating the impact of retransmission times, transceiver antenna configuration, and codebook selection on transmission reliability and parameter optimization, it is proved that the reliability requirements of URLLC scenarios can be met through reasonable configuration of the above parameters. The simulation results and conclusions in this paper can also provide reference and parameter configuration experience for the practical application and deployment of URLLC.

Abstract：The diabetic patient group urgently pursues a high-quality life which can accurately control blood glucose, and urgently needs blood glucose monitoring technology with high precision and comfortableness. However, the traditional finger blood method can only achieve a single invasive blood glucose monitoring, which can no longer meet the demand. As an important development direction of next-generation for blood continuous glucose monitoring (CGM) technology, wearable flexible blood glucose sensor shows great potential application prospects in diabetes management, and is hot research topic. This paper reviews the research progress of wearable flexible and CGM technologies in recent years, discusses the challenges and possible breakthroughs of the wearable flexible CGM technologies, and finally prospects the trend of wearable flexible blood glucose sensor commercialization.

Abstract：According to the issue of limited and low-effective peripheral bandwidth in transmission system, this paper designs a DMA high-speed transmission system based on PCIe. Taking FPGA as control unit, the transaction layer protocol of full duplex DMA read and write channels is realized with the time-sharing control method which combined PIO and DMA. Then PIO receives the instructions and translates real-time status information, DMA controller takes charge of the transfer size with a two-section-slice clipping mechanism, transmit engine arbitrates all of the requests according to the priority logic, receive engine performs real-time management of completion packet by caching the TAG label, and solves out-of-order problem. At last, the transmission speed is tasted by means of clock counter， and the restriction factors affecting read/write transmission speed are analyzed. By the practical verification, the effective bandwidth of DMA write has reached 75%, which is very close to the theoretical limit of 80%, and DMA read effective bandwidth has also reached 60%. The design solved the problem of bandwidth limitation in high-speed transmission system, and affords certain engineering application value.

Abstract：The Wafer Level Chip Scale Packaging (WLCSP) of chips on 12-inch wafers without dicing poses significant technological challenges for automatic optical inspection (AOI) in terms of high-quality imaging and imaging speed. To achieve high-quality and high-speed registration of the image sequence, a local image sequence generated by multiple scans needs to be stitched together to form a global image of the wafer. To this end, a phase correlation method based on the four-neighborhood checkerboard registration is implemented using OpenCL in an FPGA to address the challenge. Initially, a two-dimensional Fast Fourier Transform (FFT) and cross-power spectrum function kernel are constructed. Then, dual-port and row-buffered device global memory are employed to reuse the computed spectral data and to apply kernel channel cascading to enhance the registration speed. Finally, the registration result is optimized using a minimum spanning tree algorithm to reduce the cumulative error of global image coordinate calculation. The proposed registration algorithm and its accelerated performance are verified using actual scanned images.

Abstract：In order to meet the needs of real-time and high-speed data acquisition cache in the field of high-speed transmission system combined with Xilinx′s XCKU060 based on Ultrascale architecture, on the basis of understanding the definitions and characteristics of the node corresponding to FPGA and DDR3, this design allocates all the pins that connect them properly, and make them run successfully on IP cores. For the convenience of users in the use of software, based on the above, read-write FIFO and read-write logic control modules are introduced to the controller interface, optimizing its interface encap sulation.The read-write process is tested in VIVADO software. The method can meet the high speed, large capacity, real-time data read-write requirements, and take advantage of the flexibility of DDR3 storage.

Abstract：This paper proposes an intelligent car navigation solution based on indoor visible light positioning technology. In terms of visible light positioning, spatial coordinates with errors below 12 cm can be obtained through fast Fourier transform (FFT), least squares method and Tolay iteration algorithm. As for intelligent car navigation, the A* algorithm is adopted, and the current spatial coordinates are obtained through visible light positioning to correct the navigation route of the intelligent car, achieving navigation accuracy of below 15 cm for more precise navigation.

Abstract：With the development of the industrial Internet, single pair Ethernet is widely used in the field of industrial IoT and intelligent manufacturing with its data transmission and remote power supply capabilities, efficient use of space and cost savings. The single pair Ethernet protocol conversion gateway designed in this paper converts the single pair Ethernet Modbus TCP protocol to MQTT protocol, and sends the data to Ali Cloud Internet of Things platform. This paper first introduces the hardware and software architecture of the single pair Ethernet protocol conversion gateway system, and then verifies its feasibility by setting up the experimental environment and deploying the EdgeX Foundry platform. The test results show that the hardware of the single pair Ethernet gateway designed in this paper can carry out data transmission, the single pair Ethernet network port can reach the theoretical transmission rate under different bandwidth and length conditions, and the software of the single pair Ethernet gateway can receive Modbus TCP data normally and realize MQTT communication.

Abstract：In NOMA downlink multi user transmission scenario, the base station transmits a data packet by multi-user superposition to a destination in the frequency domain. In this system, successive interference cancellation (SIC) receiver is used to eliminate the co-channel interference for each user. Before demodulation and SIC, a blind detection for multi user parameter is required to reduce the signaling overhead of the NOMA system. By utilizing the shape features of constellation map, a blind detection algorithm for multi parameters based on wavelet clustering is proposed, which utilizes the grid-based method in clustering. The simulation results show that the proposed algorithm achieves higher detection performance than the max-log likelihood estimation (MLLE) algorithm under the approximate computation complexity.

Abstract：A real-time velocity planning algorithm is proposed to improve the processing accuracy and efficiency of small line segments. The algorithm consists of two parts: a dynamics-based preliminary planning algorithm is proposed to optimize the actual acceleration of short line segments and improve the smoothness of motion, addressing the jitter problem of rigid-flexible coupling and heavy-load platforms in numerical control (NC) motion; a dynamics-based adaptive multi-segment look-ahead velocity planning algorithm is proposed to enhance processing efficiency for small straight lines/arcs, addressing the inefficiency issue of the single-segment look-ahead algorithm. The algorithm was tested on a PC-based real-time controller and a gantry dual-drive XY laser cutting platform and was compared with the adaptive feed-rate algorithm and single-segment look-ahead algorithm. The results show that the proposed algorithm improves both processing quality and efficiency. Additionally, the algorithm is simple and effective, and thus suitable for high real-time application environments.

Abstract：The SCADA system is often located on the control room, which is operated by the specialist operators. It will brings more convenience in the management and decision making for the administrators, if they could monitor the running diagram and the data chart of the unit in their office. In this paper, a design way on the remote browsing system based on UDP and FTP protocol is studied. It describes the architecture, the software position and the module division of the system firstly, then describes the design of the overall flow, the data interaction technology between the processes, the file download system and the shared memory in the client. The system could pick up accurate information conveniently by the simple way and has been widely applied to the multiple fields of the industrial control.

Abstract：In order to realize a high resolution, low phase noise, high spurious rejection, small size and low cost X-band frequency source, and to solve the problems of low frequency resolution of traditional phase-locked loop frequency synthesis and low output frequency of direct digital frequency synthesis, an X-band frequency source design method based on harmonic mixing and small-step phase-locked loop to improve phase noise and frequency resolution is proposed. This method adopts the architecture of harmonic mixing to output coarse step RF signal and HMC830LP6GE phase-locked loop to output fine step mixing loop, reducing the module's discrimination and fixing it to 1∶1.The results show that the X-band frequency source has an output frequency range of 8 GHz~12 GHz, an output power greater than 13 dB, a frequency step of 1 MHz, a spurious rejection better than 60 dBc, and a phase noise better than -115 dBc/Hz@10 kHz. The frequency source has the advantages of small hopping step, small size, low spurious and low phase noise, which can be applied to various applications requiring miniaturized, low phase noise and low spurious X-band frequency source.

Abstract：In the angle tracking of radar, communication, TT&C and other fields, sum-difference amplitude-comparison monopulse angle measurement technology that based on phased array antenna is the main methodology. However, the following angle measurement errors exist. One is the estimation error of signal amplitude caused by low SNR. The other is the pattern distortion error caused by changing the beam pointing. In order to reduce these errors, the cyclic-shifted autocorrelation is proposed to estimate the signal amplitude, so that the in-band noise could be mitigated. In addition, the concentric ring arrays with structural symmetry is utilized to reduce the distortion of the pattern furtherly. The implementation steps and computational complexity analysis of the new algorithm is present. Simulation illustrates that new methodology can effectively improve the precision of angle estimation.

Abstract：In order to meet the requirements of satellite mass production for testing work, an automated testing method is introduced for the time-consuming and heavy task of model flight testing. This article specifically introduces the basic architecture of model flight testing and elaborates on the model flight automated testing system, including four parts: automatic task injection, telemetry automatic interpretation, data automatic reception, and image automatic analysis. At present, the model flight automated testing system has been successfully applied to satellite testing work, achieving a leap from traditional manual testing of a single satellite to parallel automated testing of multiple satellites, thereby reducing labor costs, ensuring testing quality, and improving work efficiency.

Abstract：In order to adapt to the rapid development trend of low cost, short cycle, and mass production in domestic commercial aerospace, and in response to the current situation of low integration of various systems on the satellite, a satellite power distribution and thermal control management unit is designed with the microcontroller TMS570 as the core. Each module of the system is integrated using an internal bus, and communicates with the central computer through CAN bus for data exchange, achieving functions such as on/off control of power distribution for each component on the satellite, temperature collection and heating bands control for the satellite, power status monitoring, SADA driving, and software updating in orbit. The system has been applied to multiple satellites in the "Jilin-1" constellation, the ground and in orbit test results show that it has the characteristics of low power consumption, low cost, standard interface, high integration, and strong scalability.

Abstract：High-performance 3D graphics and image computing is of great significance for the PKS system to expand the existing ecology and solve scientific, engineering and data analysis problems. This paper focuses on the exploration of 3D graphics and image display technology in the PKS system, and studies different 3D graphics display on the PKS system. The technical route is compared, and the performance of different 3D graphics application programming interfaces is compared on two mainstream GPU platforms. The experimental results aim to provide important support and reference for the development of 3D graphics and image display technology in the future PKS system.

Abstract：Distributed database plays an important role in PKS system, its performance and stability are the key factors to improve the information technology application innovation industry. This paper focuses on the optimization research of CockroachDB based on PKS system, puts forward the performance improvement scheme of distributed database CockroachDB under PKS platform, and conducts a large number of comparative experiments on the proposed scheme. Experimental results show that the proposed scheme can improve the performance of the database significantly, the provide a solid foundation for domestic trusted computing and trusted storage.