Please wait a minute...

Current Issue

    20 August 2024, Volume 40 Issue 8
    Research and Development
    User allocation technology for multi-dimensional resource optimization of LEO satellite
    Jiaen ZHOU, Yafei ZHAO, Mugen PENG
    2024, 40(8):  1-10.  doi:10.11959/j.issn.1000-0801.2024197
    Asbtract ( 158 )   HTML ( 29)   PDF (2938KB) ( 132 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Low earth orbit (LEO) satellite communication technology can provide ground users with broad coverage, high capacity, and high reliability services, making it one of the key supporting technologies for future networks. Direct-to-mobile satellite communication technology offers advantages such as portability, low cost, and large user scalability, promising extensive commercial and application prospects. Simultaneously, it generates corresponding demand for service resources from LEO satellite. In scenarios where LEO satellite resources were constrained, a service model was proposed that comprehensively considers communication, sensing, computation, and storage. Based on this model, a user allocation method was proposed. Simulation results indicate that this strategy, compared to three other methods, achieves higher resource utilization rates, serves more mobile phone users, and provides theoretical support for the development of direct-to-mobile satellite communication technology.

    Research on fault prediction of computer network nodes driven by log information
    Yuxi WANG, Qingwei YE, Peng ZHOU, Bing LI, Xiaodong WANG
    2024, 40(8):  11-22.  doi:10.11959/j.issn.1000-0801.2024168
    Asbtract ( 122 )   HTML ( 14)   PDF (2465KB) ( 306 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    A fault prediction method driven by log information was proposed to address the impact of node failures on normal business operations in computer networks. By constructing an efficient deep learning model and introducing a correction mechanism, node failures in computer networks were predicted and diagnosed to meet the needs of network operation and maintenance. Firstly, the log information generated by each node in the computer network was collected, the state vectors of each node and the state matrices of all nodes were obtained, then the dataset through the state filling principle was supplemented, and finally the fault prediction problem into a time series prediction problem was transformed. The performance evaluation is conducted on the publicly available small-scale operation and maintenance dataset GAIA, and the experimental results show that compared with other algorithms, the proposed model has good predictive performance in local network scenarios, and its predictive effectiveness is verified, providing a certain reference value for computer network fault prediction research.

    A VVC intra coding method based on fast partition for coding unit
    Hui ZHONG, Yu LU, Haibing YIN, Xiaofeng HUANG
    2024, 40(8):  23-33.  doi:10.11959/j.issn.1000-0801.2024196
    Asbtract ( 105 )   HTML ( 8)   PDF (2893KB) ( 85 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Compared to the high efficiency video coding (HEVC) standard, the latest generation coding standard, versatile video coding (VVC) has introduced many new technologies, including quadtree (QT) and multi-type tree (MTT) partitioning. MTT partition is extended from QT partition in HEVC. The new partition method increases encoding complexity, leading to a sharp increase in encoding time. To reduce encoding complexity, a fast intra coding method combining deep learning methods and early decision in the MTT direction was proposed. Firstly, a light-weight convolutional neural network (CNN) network was used to predict partition for QT and partial MTT. Then, an early prediction for MTT partition direction method was adopted for further optimization of residual MTT. Experimental results show that the proposed method can significantly reduce encoding complexity, with a 74.3% reduction in encoding time compared to the original encoder with only 3.3% rate loss. Moreover, the performance of proposed method is superior to other comparative algorithms.

    Research on reference signal scheme for high-mobility scenarios in the Terahertz frequency band
    Liping LIU, Tong BAO, Yu XIN, Liujun HU
    2024, 40(8):  34-41.  doi:10.11959/j.issn.1000-0801.2024178
    Asbtract ( 94 )   HTML ( 7)   PDF (1518KB) ( 60 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    A novel reference signal scheme, E DFT-s-OFDM DMRS was proposed for high-mobility scenarios in the Terahertz band. The proposed scheme involved segmenting the reference signal sequence with cyclic prefix and cyclic suffix into a head and tail reference signal sequence. These sequences were then embedded into the tail and head of each OFDM symbol, respectively. The scheme had the advantage of using a complete reference signal sequence formed by the tail of the preceding OFDM symbol and the head of the following OFDM symbol. This allowed the receiver to perform channel estimation in each OFDM symbol interval, improving the accuracy and real-time performance of channel estimation. Furthermore, adjacent OFDM symbols share identical head and tail reference signal sequences, eliminating the traditional cyclic prefix for each OFDM symbol and improving spectral efficiency. Simulation results demonstrate that this scheme provides more precise channel estimation and greater spectral efficiency than DMRS scheme with 5G NR DFT-s-OFDM waveform for high-mobility scenarios in the Terahertz band.

    Lightweight face image restoration algorithm based on multi-scale feature fusion
    Xiao ZHAO, Ziyi ZHAO, Chen YANG
    2024, 40(8):  42-51.  doi:10.11959/j.issn.1000-0801.2024183
    Asbtract ( 108 )   HTML ( 12)   PDF (4357KB) ( 133 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the problems of poor quality of restored images and large number of model parameters in the current occluded face image restoration, a lightweight face image restoration model based on multi-scale feature fusion with improved U-Net, LM-UNET, was proposed. Firstly, the original convolution was replaced by a depthwise separable convolution to enhance the feature expression ability of the model for different channels and contextual information. Secondly, a multi-scale feature attention fusion module was designed in the jump connection to fully fuse the information of different scale features, and the embedded residual block reduced the semantic gap between features to improve the repair accuracy of the model. Finally, a positional attention module was introduced to enhance the salient information of the face image, and improve the model’s effective extraction ability of facial positional pixel information of the model. The algorithm was trained, validated and tested on the occluded face dataset MFD generated based on the CK+ dataset, and the PSNR of the repaired image reached 30.49 dB and SSIM reached 96.85%. The experimental results of comparing the model with the other models show that the model has better image quality and visual effect for restoration of the face in the presence of occlusion.

    Adaptive distributed cloud edge collaborative load control strategy for load management
    Siwei LI, Li JIN, Long YU, Lishi DU, Liang YUE, Xirun ZHANG
    2024, 40(8):  52-62.  doi:10.11959/j.issn.1000-0801.2024192
    Asbtract ( 74 )   HTML ( 5)   PDF (2351KB) ( 121 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    To solve the problem that the controllable load management of multiple controllable load resources requires a lot of computing resources and can not achieve accurate automatic power control, a cloud-edge cooperative load resource allocation strategy for multiple controllable load regulation was proposed. Firstly, the collaborative control framework of cloud edge was designed to integrate and process the data of various controllable load resources. Secondly, considering the similarity of computing tasks of different edge nodes, the optimization goal was to minimize the time cost of all computing tasks, and the cloud computing resource allocation strategy was given to allocate computing resources reasonably. Finally, the computational resource allocation was solved by genetic algorithm based on adaptive cross-mutation probability. Finally, the calculation of resource allocation was solved using a genetic algorithm based on adaptive crossover mutation probability. The experimental results show that the algorithm proposed has significant advantages in task completion time and execution cost, and these advantages become more pronounced as the number of tasks increases and computing resources decrease. It can significantly improve computing efficiency and reduce computing time.

    Deterministic service level agreement: indicator classification in private networks for vertical industry
    Mufeng ZHANG, Hongxing LI, Ke WANG, Xiaoliang LI, Yaqiong LIU, Yihong HU, Guochu SHOU
    2024, 40(8):  63-77.  doi:10.11959/j.issn.1000-0801.2024182
    Asbtract ( 101 )   HTML ( 11)   PDF (1808KB) ( 201 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Meeting the differentiated deterministic service quality experience of customers in different industries in an economically effective manner is conducive to promoting the construction of a vertical industry deterministic private network. In the face of various segmented industries and different dimensions of deterministic service quality requirements, the classification and grading of SLA(service level agreement) indicators are very important. Firstly, various types of application scenarios of deterministic service in vertical industries were summarized, and indicator requirements analysis was conducted for each type of business. Then, a deterministic SLA indicator system was proposed and a hierarchical design scheme for each indicator was summarized to solve the problem of insufficient reflection of deterministic attributes in current network SLA indicators, and help reducing the difficulty of understanding and aligning indicators between industry customers and service providers, and enable the network to meet the deterministic service needs of different industries. Finally, prospects for future work were presented.

    Dynamic heterogeneous network representation learning method based on Hawkes process
    Lei CHEN, Kun DENG, Xingyan LIU
    2024, 40(8):  78-93.  doi:10.11959/j.issn.1000-0801.2024195
    Asbtract ( 56 )   HTML ( 2)   PDF (3121KB) ( 71 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Existing methods for heterogeneous network representation learning mainly focus on static networks, overlooking the significant impact of temporal attributes on node representations. However, real heterogeneous information networks are very dynamic, and even minor changes in nodes and edges can affect the entire structure and semantics. In this context, a dynamic heterogeneous network representation learning method based on Hawkes process was proposed. Firstly, the vector representation of nodes was obtained by utilizing the relational rotation encoding method and attention mechanism, where the attention coefficients of adjacent nodes were learned. Secondly, the optimal weighted combination of different meta-paths was learned to better captures the structural and semantic information of the network. Finally, leveraging the time decay effect, time features were introduced into node representations through the formation of neighborhood sequences, resulting in the ultimate embedding representation of nodes. Experimental results on various benchmark datasets indicate that the proposed method significantly outperforms baseline methods. In node classification tasks, Macro-F1 average is increased by 0.15% to 3.45%, and NMI value in node clustering tasks is improved by 1.08% to 3.57%.

    Performance evaluation and system design of 6G cooperative integrated sensing and communication technology
    Peng GAO, Feifei ZHOU, Qixing WANG, Lixiang LIAN, Jinpei YU, Tao MA
    2024, 40(8):  94-107.  doi:10.11959/j.issn.1000-0801.2024199
    Asbtract ( 235 )   HTML ( 38)   PDF (5315KB) ( 294 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    As a promising technique for 6G integrated sensing and communication, cooperative integrated sensing and communication (CoISAC) technology harnesses additional sensing information from various network nodes to overcome challenges of ISAC systems such as low sensing accuracy, non-line-of-sight path scenario, and limited sensing range. The current status of CoISAC technology was summarized and analyzed. A novel performance metric and two cooperative approaches were proposed. The challenges faced by CoISAC technology were analyzed, and the future development trend of CoISAC technology was forecasted.

    Low-carbon optimization planning method for integrated energy system based on DG uncertainty affine model
    Tao JIANG, Cong XU, Shaohui JIA, Shen WANG, Yajian ZHANG
    2024, 40(8):  108-120.  doi:10.11959/j.issn.1000-0801.2024162
    Asbtract ( 61 )   HTML ( 1)   PDF (2909KB) ( 53 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the problem that the output of distributed generators (DG) of new energy sources such as wind power and photovoltaic is uncertain due to changes in environmental factors, and the transaction price of the existing carbon trading model is fixed, resulting in increased carbon reduction costs, an integrated energy system optimization planning method that takes into account dynamic carbon emission constraints and DG uncertainty was proposed. Firstly, a DG output model based on matrix affine algorithm was established according to environmental conditions to reduce the impact of DG output uncertainty on the optimization planning of the integrated energy system. Secondly, carbon emissions was introduced as a punitive measure into the optimization planning of the integrated energy system to improve the traditional carbon trading model and reduce the carbon emissions of the integrated energy system. Then, based on the differential evolution-particle swarm optimization algorithm, the established low-carbon planning model of the integrated energy system was solved to avoid the algorithm from falling into local optimality during the optimization process. Finally, the simulation results on an IEEE 33 node system show that the proposed planning method reduces the total investment cost by 8.68% and 2.93% respectively compared with the traditional stochastic optimization and interval optimization planning methods. Compared with the traditional fixed carbon trading price model, carbon emissions are reduced by 6.28%.

    Deep learning based demodulation method for multiple access of visible light communication in 6G
    Xinyu SHAO, Yao YAO, Liang WANG
    2024, 40(8):  121-129.  doi:10.11959/j.issn.1000-0801.2024203
    Asbtract ( 92 )   HTML ( 6)   PDF (3015KB) ( 77 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Because the modulation bandwidth of the light emitting diode is narrow, the capacity of the visible light communication system is limited, the spectral efficiency and number of endpoint user can be improved by the multiple access technique. However, there is strong inter-user interference among multiple access users in the visible light communication system. In view of this problem, by utilizing the correlation among received signals of the visible communication system, a multiple user detection and signal recovery method of multiple users for multiple access based on deep neural network was proposed. The transmitter model and the receiver model of the visible light communication system were presented based on sparse code multiple access, the temporal convolutional network was adopted to learn the inter-signal temporal correlation of the long sequence, the output sequence was delivered to dense layer to learn the spatial mapping relationship of the signal sequence, in the end, signals of all users were recovered in the receiver of the visible light communication system. Experimental results indicate that the proposed signal recovery method improves the communication performance of the visible light communication system multiple access effectively, and the proposed method can play an active role under the condition of different communication distances, different signal noise ratios and transmit speeds.

    Construction and application of a quantitative efficiency assessment model for green AI in intelligent customer service systems
    Xiaoliang MA, Ying LIU, Ruqiang ZHAO, Bangxing YANG, Jie GAO, Congjian DENG
    2024, 40(8):  130-137.  doi:10.11959/j.issn.1000-0801.2024184
    Asbtract ( 103 )   HTML ( 5)   PDF (1292KB) ( 102 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The rise of AI and the increasing prominence of LLM are positioned as core technologies for knowledge dissemination and multi-turn dialogues. Alongside their growth, the high energy consumption associated with data processing, model training, and deployment in AI large models is necessitating effective evaluation to facilitate quantitative comparisons before and after model optimization. An assessment method for the energy consumption of AI large models was introduced, aimed at quantitatively evaluating the service efficiency (E) of AI models. This model was incorporated with multiple dimensions such as training convergence time (T), model parameter size (P), and floating-point operations (F), and quantitative analysis was achieved through the construction of an energy consumption function C(T, P, F). Furthermore, by employing the nonlinear least squares method, model parameters were derived. This analysis method was not only applicable to the operational efficiency analysis of AI models used by telecommunications operators but can also be generalized for energy consumption assessment of AI models across various industries.

    Engineering and Application
    Research on the construction of standard digital capability maturity model for grid enterprise
    Haiqing XU, Sining WANG, Zhifeng WEI, Bingqiang GAO, Baobing XIA, Zhenxia ZHAO
    2024, 40(8):  138-148.  doi:10.11959/j.issn.1000-0801.2024198
    Asbtract ( 144 )   HTML ( 8)   PDF (2414KB) ( 68 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The digitization of standards is increasingly becoming a core trend in both international and domestic development. To achieve transformation and upgrading, enterprises need to scientifically analyze and evaluate their capabilities in standard digitization. Current research has primarily focused on two areas: the maturity model for enterprise digital transformation and the general model for standard digitization capability levels. However, there is still a lack of research on the maturity evaluation model for standard digitization capabilities specifically tailored to the practical business operations of specific industries or companies. By drawing on the concept of capability maturity model (CMM), combining the full life cycle management of standards and the implementation of technical standards in power grid enterprises, and carrying out expert interviews, a well-structured standard digitization capability maturity model was constructed, which consisted of two major dimensions, namely, five standard digitization capability domains and seven standard digitization process domains. Taking the business expanding scenario of a certain company of the State Grid as an example, a two-dimensional table evaluation method was used to get the standard digitization capability of the company in this business scenario to reach the L2 exploratory level, which verifies the practical application effect of the model.

    Research on intelligent level of the brain of computility network
    Chen WANG, Peng ZHAO, Guohua TANG, Xiaohan WANG, Xiangyang YUAN
    2024, 40(8):  149-161.  doi:10.11959/j.issn.1000-0801.2024153
    Asbtract ( 101 )   HTML ( 9)   PDF (1999KB) ( 181 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Brain of computility network is the intelligent center of computility network, and it is the key system to realize multi-element orchestration and scheduling of computility network. Intelligence is the core of realizing and improving the flexible, agile and efficient orchestration, management and operation capability of the brain of computility network. Many companies in the communication industry have focused on the introduction of intelligent technology of the brain of computility network on the basis of the early realization of the products, and put forward high expectations for the intelligent evolution of the brain of computility network in the future. However, at present, there is no systematic evaluation method for the intelligence level of the brain of computility network, which cannot well evaluate the intelligence level, upgrading stages and key indicators to be optimized. Based on the existing work and demand analysis in the industry, the goal and stage capability of the intelligent evolution of the brain of computility network were studied, and a classification method of the intelligent level of the brain of computility network was put forward, which included detailed steps such as the selection of the evaluation object, the decomposition of the evaluation process, the consideration of the constraint indicators, and a complete set of rules for the evaluation of the intelligent level of the brain of computility network. Finally a concrete application example was given. The proposed method can effectively quantify the specific level of the brain of computility network system, clarify the direction and index of its evolution to the next level, and provide reference for industry intelligent evaluation and system construction.

    Research on multi-party security collaborative linear regression for computing power networks
    Jie PAN, Huifang HOU, Xi CHEN, Zhao XUE, Liankun XU
    2024, 40(8):  162-171.  doi:10.11959/j.issn.1000-0801.2024204
    Asbtract ( 64 )   HTML ( 3)   PDF (2579KB) ( 72 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the rapid development of science and technology, machine learning has become a key factor driving the progress of enterprises. However, for small and medium-sized enterprises, the amount of data and computing power often become obstacles to apply machine learning. The rise of computing power networks has brought new opportunities for enterprises, accompanied by new challenges such as data security. A linear regression scheme for multi-user security collaborative computing based on computing power networks was proposed. The scheme allowed multiple users to achieve secure joint training in a computing power network using sensitive data to build a linear regression model. The scheme adopts low-cost blinding method and homomorphic encryption technology to encrypt user sensitive data to protect the security of sensitive data.

    5G deterministic network key technology research and application
    Shiyin ZHU, Jian YANG, Lei KONG
    2024, 40(8):  172-182.  doi:10.11959/j.issn.1000-0801.2024154
    Asbtract ( 148 )   HTML ( 29)   PDF (2293KB) ( 123 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    5G deterministic network is an important part to support the construction of 5G+ Industrial Internet, and it is of great significance to study the key technologies and landing schemes of 5G deterministic network. Firstly, 5G ultra-reliability low latency communication technology and frame replication and cancellation technology were studied to reduce transmission latency and improve reliability. Secondly, cyclic queue forwarding technology was explored in 5G networks to eliminate end-to-end jitter. Finally, a 5G deterministic network landing scheme was proposed and tested in actual industrial production labs. The test results show that the comprehensive performance of the proposed network scheme is better than the existing network schemes, and can meet the demand for 5G deterministic networks in many industrial scenarios.

Copyright Information
Authorized by: China Association for Science and Technology
Sponsored by: China Institute of Communications
Posts and Telecom Press Co., Ltd.
Publisher: Beijing Xintong Media Co., Ltd.
Editor-in-Chief: Chen Shanzhi
Editorial Director: Li Caishan
Address: F2, Beiyang Chenguang Building, Shunbatiao No.1 Courtyard, Fengtai District, Beijing, China
Postal Code: 100079
Tel: 010-53879277
        010-53879278
        010-53879279
E-mail: dxkx@ptpress.com.cn
Mailing Code: 2-397
ISSN 1000-0801
CN 11-2103/TN
Visited
Total visitors:
Visitors of today:
Now online: