Subjects: Computer Science >> Computer Application Technology submitted time 2023-02-15 Cooperative journals: 《桂林电子科技大学学报》
Abstract: With the advent of Digital Journal publishing, online journal editing management system is used to collect, submit,
distribute, edit, review, expert review, finalize, query the results of manuscript processing, keyword and reference review,
layout settings, distribution and typesetting, three-level issuance, typesetting and proofreading, sample production, data
statistics, cost management, year-end index and other functions. Realize the process control of journal editing and review,
and improve the reliability and transparency of journal article editing and review. This paper introduces the process of system
implementation, including requirement analysis, system design, database design, system implementation and system testing.
The system has been put into use and the system is in normal use.
Subjects: Computer Science >> Computer Application Technology submitted time 2023-02-15 Cooperative journals: 《桂林电子科技大学学报》
Abstract: In order to verify whether the given states satisfiy the bisimulation, a local algorithm of fuzzy bisimulation is proposed.
The algorithm takes verification and traversal at the same. While verify the given states whether satisfiy the fuzzy
bisimulation, the state space is dynamically increased, so that the algorithm only needs to traverse part of the state space to
complete the verification. In some cases, the local algorithm of fuzzy bisimulation can verify whether the given states satisfiy
the fuzzy bisimulation more quickly, especially when the two states do not satisfy the fuzzy bisimulation. The local algorithm
and the existing global algorithm are implemented by Java and compared by experiments.The experiments,shows that
this algorithm is more efficient than the existing global fuzzy bisimulation algorithm when the given states do not satisfiy the
bisimulation.
Subjects: Information Science and Systems Science >> Basic Disciplines of Information Science and Systems Science submitted time 2023-02-14 Cooperative journals: 《桂林电子科技大学学报》
Abstract: Aiming at the structure of the lightweight authentication encryption algorithm ASCON, a differential power analysis)
method is proposed. It combines the implementation characteristics of the algorithm S-box, uses the Hamming weight
model as the power consumption discrimination function, groups the traces, and recovers the master key for encryption.
Furthermore, for the "ghost peaks" what appear in DPA attacks, a traces preprocessing method is given. First, the traces
are grouped according to plaintext and averaged, and then DPA attacks are launched on the preprocessed traces. The 44 bit
master key of ASCON cipher can be recovered by attacking its sa permutation, where 1 500 traces are collected. In addition,
the time required to directly attack the original traces is 21 849.888 9 ms, and the time required to attack the preprocessed
traces is 198.911 3 ms. After preprocessing the traces, the time taken to attack the preprocessed traces is about 1/109 of
that of directly attacking the original traces.
Subjects: Computer Science >> Computer Application Technology submitted time 2022-11-02 Cooperative journals: 《桂林电子科技大学学报》
Abstract: Subgraph matching is a fundamental problem in graph data analysis and has important research significance. Aiming
at the problem of a large number of redundant searches in the subgraph matching algorithm, a subgraph matching
symbol algorithm based on graph neural network(SSMGNN) was proposed. The algorithm used the graph neural network
technology to aggregate the neighborhood information of nodes, and obtained the feature vector containing the local attributes
and structure of the graph, and used the vector as the filter condition to obtain the node candidate set C of the query
graph. In addition, optimizing the matching order and using symbolic ADD operations to construct each candidate region of
C in the data graph reduced redundant searches during subgraph enumeration verification. The experimental results show
that, compared with the VF3 algorithm, the algorithm effectively improve the solving efficiency of subgraph matching.
Subjects: Computer Science >> Computer Application Technology submitted time 2022-11-02 Cooperative journals: 《桂林电子科技大学学报》
Abstract: The data distribution algorithm in Ceph distributed system only takes the capacity as the standard for selecting
storage nodes, and does not consider the network state and node load of storage nodes. In the replica mode, when a storage
node in the three replicas needs to be repaired, too high node load or network load will lead to a large node repair delay. To
solve this problem, a fault-tolerant node selection algorithm based on Ceph (FTNSC) is proposed. Firstly, the software defined
network technology is used to obtain the real-time network state and node load information as the data support of the
node selection method; Then, a multi-attribute decision-making mathematical model considering the node load information is
established to determine the location of the primary storage node; Finally, the artificial bee colony algorithm is used to obtain
the optimal secondary storage node according to the network state and node performance with the primary storage node.
Experimental results show that compared with the existing methods, the proposed method can improve the performance of
data storage nodes and reduce the data repair delay when nodes fail.
Subjects: Computer Science >> Computer Application Technology submitted time 2022-11-02 Cooperative journals: 《桂林电子科技大学学报》
Abstract: Virtual machine placement is a key step in the process of virtual machine consolidation. The quality of the virtual
machine placement method usually affects the resource utilization efficiency and performance of the cloud data center. Such
problems can be solved by establishing a multi-objective optimization model. Currently, cloud data centers have high energy
consumption, low resource utilization, and resource fragmentation. In view of the above situation, a virtual machine placement
strategy based on MALO algorithm is proposed. By establishing a multi-objective and multi-constrained virtual machine
placement model, the energy consumption, resource utilization, and resource fragmentation are optimized. And on the
basis of the Antlion algorithm, by improving the boundary change strategy of the solution space and the location selection
strategy of ants random walk, finally the position of the ants is corrected beyond the boundary, so that the diversity of the
population can be better guaranteed, which can better Jump out of the local optimal solution. Based on the virtual machine
placement platform, the simulation experiments of MALO algorithm and four other virtual machine placement algorithms
are carried out. The experimental results show that compared to the Antlion algorithm, BRC algorithm, MBFD algorithm
and FFD algorithm, the MALO algorithm has a certain improvement effect in reducing energy consumption, improving resource
utilization and reducing resource fragmentation.
Subjects: Computer Science >> Computer Application Technology submitted time 2022-11-02 Cooperative journals: 《桂林电子科技大学学报》
Abstract: In order to improve the reconstruction efficiency of the very large-scale integrated logic array, an improved algorithm
based on the principle of first selecting the shortest and long interconnect is proposed. This improved algorithm constructs
logical columns from both ends of the array until the two logical columns intersect, then stops constructing logical
columns. In the local area with two intersecting logical columns as the boundary, the processor unit with the shortest link
length of each row is searched from top to bottom, and the selected processing unit is used to construct the locally optimal
logical column. Based on the above operations, using the idea of divide and conquer, the newly obtained logical column is
used as the boundary of the new local area, and the new local optimal logical column is obtained in turn. Finally, connect the
obtained local optimal logical columns to obtain the final target array. Through comparison and analysis with existing reconstruction
algorithms, the efficiency of the proposed algorithm is verified. The final simulation results show that under the
condition that the scale of the logic array remains unchanged, compared with the existing reconstruction algorithm, this
improved algorithm can effectively reduce the number of processor accesses during the array reconstruction process, and to a
certain extent, the running time of the reconstruction is reduced, and the reconstruction efficiency of the logic array is improved
Subjects: Information Science and Systems Science >> Basic Disciplines of Information Science and Systems Science submitted time 2022-10-26 Cooperative journals: 《桂林电子科技大学学报》
Abstract: To solve the problem that agents cannot make effective decisions under local observation, a conflict resolution
method combined with deep reinforcement learning is proposed. Based on DDQN algorithm, this method uses the characteristics
of reinforcement learning mode to calculate the cumulative return of agent and determine the priority of agent through
the return value, so as to achieve the purpose of conflict resolution. The method is evaluated by simulating the traffic jam in
real life, and the experimental results show that the method can effectively solve the agent conflict.
Subjects: Information Science and Systems Science >> Systematic Application of Information Technology submitted time 2022-10-26 Cooperative journals: 《桂林电子科技大学学报》
Abstract: In order to solve the data matching problem between sports camera and lidar, a joint calibration optimization method
of sports camera and lidar based on Levenberg-Marquard (LM) algorithm is designed. First, the calibration board is
placed in the common field of view of the lidar and the sports camera, and the laser point cloud and image data of the calibration
object at different positions are collected by changing the position of the calibration board. Then the fisheye distortion
correction function is called through OpenCV to correct the image distortion, and obtain multiple sets of pixel coordinates of
the corner points of the calibration plate image. At the same time, point cloud filtering and point cloud registration are performed
on the laser point cloud, and the laser point cloud is segmented by a combination of manual and automatic methods,
and then the point cloud center iterative algorithm is used to solve the calibration board point cloud center coordinates and
The point cloud coordinates of each corner point. Finally, through multiple sets of point cloud coordinates representing the
corner points of the calibration board and the corresponding image pixel coordinates, the direct linear transformation method
(DLT) is used to calculate the initial value of the joint calibration between the two sensors, and the difference between the
point cloud reprojection coordinates and the image pixel coordinates is constructed. The least squares function of, the function
is optimized by the LM algorithm that introduces the damping factor, and the optimized joint calibration result is
solved. Experiments show that the joint calibration result reduces the reprojection error by 35% compared with the initial
value. The joint calibration result is used to achieve laser point cloud and image fusion based on the principle of collinear equations, which verifies the accuracy and effectiveness of the method
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-01-28 Cooperative journals: 《计算机应用研究》
Abstract: Aiming to solve the problem of abnormal holes occur in 3D reconstruction mesh models frequently, this paper proposed a novel method that hole-defect identification automatically and detail-preserving surface repair around the hole. Firstly, in order to determine whether the hole is a real defective hole, it detected the context irregularity and contour irregularity of the missing region. After that, it detected and matched the feature lines around the hole, which were to form the fundamental surface of the hole area, and then it introduced unconstrained triangulation for basal surface. Finally, it made use of the anisotropy of mesh to refine and adjust morphologically, which were to improve the topological and geometric properties of the mesh. The experimental results show that the proposed method can identify the defective hole and restore the hole area’s detail features of triangular mesh models effectively.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-01-28 Cooperative journals: 《计算机应用研究》
Abstract: In order to improve the efficiency of data scheduling and replica access in cloud computing, the paper studied the replica placement problem in replica strategy and proposed a replica placement strategy based on ant colony algorithm. According to the principle of ant colony foraging in nature, ant colony algorithm applies to the entire process of replica placement. The pheromone dynamic updating and Laplace probability distribution improves the ant colony algorithm, a set of optimal solutions is obtained for replica placement. Simulation is on the CloudSim platform. The experimental results show that the proposed scheme is superior to the original ant colony algorithm in average job completion time, network utilization and load balance, and reduces the copy placement time consumption and network load to a certain extent.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-10-11 Cooperative journals: 《计算机应用研究》
Abstract: Computer poetry generation is our first step towards computer writing. At present, there are many problems in computer poetry writing, such as unclear theme, the content of poetry is inconsistent with the writing intention. For improving these problems, this paper follows the process of writing poem by the ancient Chinese poet and proposes a method for generating Chinese poetry with two stages. The first stage extracts the outline. During this process, this paper use TextRank algorithm to extract keywords from user input text, and propose an attention-based sequence to sequence neural network model for expanding keyword. The second stage generates each line of poem based on the outline of poem. During this process, this paper proposed a sequence to sequence neural network model with dual-encoding and attention mechanism for generating poem. At the end, this paper verified the effectiveness of our approach by evaluation. Compared with baseline approach, the theme of the poem generated by our approach is more explicit, and the contents expressed by the poem are more consistent with the writing intention.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》
Abstract: This paper proposed a method in English sentence compression based on Pre-readin” and Simple Attention Mechanism. On the basis of Gated Recurrent Unit (GRU) and Encoder-Decoder, this paper modeled the original sentence semantics twice in the encoding stage. The first result was used as a global information to strengthen the second semantic model, thus obtaining a more comprehensive and accurate semantic vector. With full consideration of the particularity of the deleted sentence compression, this paper simply adopt the 3t-Attention mechanism in the decoding stage to improve the efficiency and accuracy of prediction, which means that the semantic vectors most relevant to the current decoding time step are inputted to the decoder. The results from the experiments on the Google news sentence compression dataset show that our model significantly outperforms all the recent state-of-the-art methods. Therefore, "Pre-reading" and Simple Attention Mechanism can effectively improve the accuracy of English sentence compression.