Current Issue

20 June 2023, Volume 39 Issue 6

Previous Issue   

Review

Data center networks with performance guarantee: a survey

Kaihui GAO, Dan LI

2023, 39(6):  1-21.  doi:10.11959/j.issn.1000-0801.2023125

Asbtract ( 258 )   HTML ( 87)   PDF (1681KB) ( 483 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

As an important information infrastructure, data center network (DCN) supports numerous distributed applications, such as artificial intelligence training and cloud storage.These applications require the transmission of large amounts of data, and as a result, the demand for performance stability of DCN has been increasing.The performance guarantee of DCN has gained widespread attention from both academia and industry in recent years.Firstly, the main challenges of implementing performance guarantees in DCN were analyzed, and the research directions to improve performance stability were proposed.Secondly, three essential capabilities for a performance-guaranteed DCN were summarized, such as high availability, bandwidth guarantee, and bounded latency.The comparative analysis of the relevant research was provided from multiple perspectives.In the end, the future development trend of the research on the performance guarantee of DCN was looked forward to.

Topic: Polymorphic Network

Design on the full-dimensional reconfiguration polymorphic network switching chip architecture

Yu LI, Zhaozhao LI, Ping LYU, Qinrang LIU

2023, 39(6):  22-32.  doi:10.11959/j.issn.1000-0801.2023129

Asbtract ( 174 )   HTML ( 35)   PDF (2939KB) ( 175 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

At present, IP network has the problems of rigid structure, poor scalability and poor security.To solve these problems, the concept of polymorphic network has been proposed by researchers and gradually become a research hotspot.The programmable network switching chip is the basis of polymorphic network implementation.Therefore, a full-dimensional reconfigurable polymorphic network switching chip architecture was proposed.Based on eFPGA in the chip, the port-level and bit-level fine-grained reconfigurable polymorphic network was realized.Based on self-designed process element (PE), the coarse-grained reconfigurable network switching engine was realized.Thereby the full-dimensional reconfigurable polymorphic network switching chip was realized and laid the foundation for the data layer of polymorphic network.To improve the network switching performance, an improved bit vector algorithm was proposed, which could realize fast flow table lookup with a smaller resource consumption.The proposed algorithm has showed its practicability in the research.

Research on incremental deployment mechanism of network modality for polymorphic network environment

Jiong LI, Yuxiang HU, Pengshuai CUI, Le TIAN, Yongji DONG

2023, 39(6):  33-43.  doi:10.11959/j.issn.1000-0801.2023123

Asbtract ( 127 )   HTML ( 18)   PDF (1978KB) ( 128 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Polymorphic networks support diverse network requirements and can coexist, evolve, and change on a unified physical or logical infrastructure in the form of network modality.They can maximize the innovation vitality of network technology driven by new applications.The evolution and development of modality require the incremental deployment of modality to meet the needs of convenience and privacy protection in the multi-user deployment process.However, the deployment of current network functions still uses the one-by-one deployment method, resulting in high complexity of the deployment of new modality, and mutual exposure of processing logic between different users.To address these issues, an incremental deployment mechanism for polymorphic network environments was proposed.Firstly, an incremental deployment mechanism for network modality was designed.Then, the incremental deployment of modality was abstracted as a graph merging problem based on this mechanism, and an adjacency list-based merge parsing graph algorithm was designed.The simulation results indicate that the proposed mechanism can achieve fast and correct merging of network modality, support incremental modality deployment, and greatly reduce the application deployment and service provision threshold of new network systems.

Network modal innovation for distributed machine learning

Zehua GUO, Haowen ZHU, Tongwen XU

2023, 39(6):  44-51.  doi:10.11959/j.issn.1000-0801.2023128

Asbtract ( 191 )   HTML ( 21)   PDF (2020KB) ( 250 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Distributed machine learning, as a popular computing architecture for artificial intelligence, still faces challenges of slow model training and poor data performance transmission.Traditional network modalities were un able to meet the communication needs of distributed machine learning scenarios, hindering the improvement of model training performance.New network modalities and operation logic for distributed machine learning scenarios using multimodal network technology were proposed.This approach was designed based on application characteristics and provides implications for the use of multimodal network technology in various industries.

Research on isolated-forwarding technology oriented polymorphic network

Aihua LIU, Hanguang LUO, Jianzhong WEN, Zhiguo ZHAN

2023, 39(6):  52-60.  doi:10.11959/j.issn.1000-0801.2023126

Asbtract ( 146 )   HTML ( 33)   PDF (1831KB) ( 278 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The polymorphic network adopted the idea of “separation of network technology system and supporting environment”, and through innovative network development paradigms, various existing or future network technology systems dynamically load and run on a fully dimensional and definable network environment in the form of network polymorph.The “self-directed and self-sustainable development” of diversified network technology systems was realized in the same physical facility.Based on the study of polymorphic network architecture, the technology of mutual isolation between network polymorphs was mainly proposed in polymorphic forwarding processing, and in the current Ethernet physical facilities, the Ethernet physical layer technology was used to realize the hard isolation function of polymorphic network forwarding, so as to achieve the effect of physical isolation of polymorphic network in the same Ethernet interface and switching medium, then provided independent forwarding environment support for network polymorph.

Research on polymorphic network application technology oriented intelligent manufacturing

Tao ZOU, Huifeng ZHANG, Wanxin GAO, Qi XU, Congqi SHEN, Jun ZHU, Zhongxia PAN, Xingchang GUO

2023, 39(6):  61-72.  doi:10.11959/j.issn.1000-0801.2023131

Asbtract ( 145 )   HTML ( 20)   PDF (8162KB) ( 97 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The rise of intelligent manufacturing has induced many new demands on network technology.The polymorphic network separates the network technology from the support environment, supporting the coexistence as well as flexible loading and operation of various network technologies in the same network support environment, and can provide diversified, customizable, and highly reliable network service capabilities.The network environment requirements of intelligent manufacturing were studied, and smart factories were taken as an example to design efficient technical solutions based on the polymorphic network for various application scenarios.Different network technologies were efficiently adapted to application scenarios such as collaborative research and development design, production line control, machine vision quality inspection, unmanned intelligent inspection, accident emergency response, etc.The study verified the technical advantages of applying polymorphic networks to intelligent manufacturing, which could provide strong support for promoting the transformation and upgrading of the manufacturing industry.

Research on intent-based cooperative architecture of polymorphic network and service

Jie ZHANG, Yongjian ZHAO, Dongrui XIAO, Yong XU, Jianhong XU, Jianjian YANG

2023, 39(6):  73-84.  doi:10.11959/j.issn.1000-0801.2023133

Asbtract ( 137 )   HTML ( 19)   PDF (2430KB) ( 145 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Different network techniques are developing autonomously within their own region, besides, various new network techniques come out increasingly.However, the different network techniques cannot coordinate with each other to support the various business efficiently.Based on polymorphic networks and business environments, the researches about the coordinated scheduling of networks and businesses were illustrated from the perspective of the operator’s management and operation.The intent-based cooperative architecture of polymorphic network and service was proposed, which was driven by polymorphic business services.Furthermore, the way to orchestrate the network capabilities intelligently, and to promote the business innovation were explicated.

Research and Development

Synthetic speech detection method using texture feature based on circumferential local ternary pattern

Honghui JIN, Zhihua JIAN, Man YANG, Chao WU

2023, 39(6):  85-95.  doi:10.11959/j.issn.1000-0801.2023121

Asbtract ( 86 )   HTML ( 13)   PDF (1323KB) ( 110 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to further improve the accuracy of synthetic speech detection, a synthetic speech detection method using texture feature based on circumferential local ternary pattern (CLTP) was proposed.The method extracted the texture information from the speech spectrogram using the CLTP and applied it as the feature representation of speech.The deep residual network was employed as the back-end classifier to determine the real or spoofing speech.The experimental results demonstrate that, on the ASVspoof 2019 dataset, the proposed method reduces the equal error rate (EER) by 54.29% and 2.15% respectively, compared with the traditional constant Q cepstral coefficient (CQCC) and linear predictive cepstral coefficient (LPCC), and reduced the EER by 17.14% compared with the local ternary pattern(LTP) texture features.The CLTP comprehensively takes into account the differences between the central and peripheral pixels in the neighborhood and between each peripheral pixel.Then it can acquire more texture information from the speech spectrogram, and improve the accuracy of synthetic speech detection.

Airworthiness analysis of low earth orbit satellite communication technology for UAV application

Yaoyu HE, Chao ZHANG

2023, 39(6):  96-104.  doi:10.11959/j.issn.1000-0801.2023117

Asbtract ( 181 )   HTML ( 47)   PDF (1280KB) ( 311 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

With the rapid development of unmanned aerial vehicle (UAV) and low earth orbit (LEO) satellite technology, the integrated network consisted of the space, air, ground and sea has promoted the wide development of Internet of remote things (IoRT) on the UAV platform.Confronting the application demand of UAV with LEO satellite, the corresponding flight safety of UAV needs to be considered firstly, and the airworthiness of UAV has become the top priority of the technical application.The significance of the airworthiness of the UAV system was introduced and the technology of LEO satellite with UAV was summarized.Moreover, the related airworthiness programs of UAV equipped with LEO communications system were proposed.Meanwhile, based on the comprehensive analysis of UAV oriented LEO satellite technology, the means of compliance (MOC) on the airworthiness theory has been carried out.

A progressive growing of conditional generative adversarial networks model

Hui MA, Ruiqin WANG, Shuai YANG

2023, 39(6):  105-113.  doi:10.11959/j.issn.1000-0801.2023134

Asbtract ( 104 )   HTML ( 9)   PDF (3261KB) ( 71 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Progressive growing of generative adversarial networks (PGGAN) is an adversarial network model that can generate high-resolution images.However, when the categories of samples are unbalanced, or the categories of samples are too similar or too dissimilar, it is prone to produce mode collapse, resulting in poor image generation effect.A progressive growing of conditional generative adversarial networks (PGCGAN) model was proposed.The idea of conditional generative adversarial networks (CGAN) was introduced into PGGAN.Using category information as condition, PGGAN was improved in two aspects of network structure and mini-batch standard deviation, and the phenomenon of model collapse in the process of image generation was alleviated.In the experiments on the three data sets, compared with PGGAN, PGCGAN has a greater degree of improvement in inception score and Fréchet inception distance, two evaluation indicators for image generation, and the generated images have higher diversity and authenticity; and PGCGAN multiple unrelated datasets can be trained simultaneously without crashing, and high-quality images can be produced in datasets with imbalanced categories or data that are too similar and dissimilar.

Big data classification method of non relational distributed submission information under differentiated requirements

Lu HAN, Weiyu CHEN, Fei ZHANG, Jianfeng HE, Huaizhen SU

2023, 39(6):  114-121.  doi:10.11959/j.issn.1000-0801.2023122

Asbtract ( 58 )   HTML ( 5)   PDF (1357KB) ( 78 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The classification method of non-relational distributed submitted information big data under the differentiated demand was studied, aiming at the problem of multi-source heterogeneous, widely distributed submitted information with more differentiated application requirements and inability to distinguish the available information.Firstly, the usability, openness and expansibility of the non-relational distributed submission information database were analyzed.The unstructured database storage TRIP was used to store non-relational distributed submission information by combining the basic requirements of field types.Then, the hashing process within the Hamming hash family was analyzed.Under the constraint of linearity level requirements, cellular automata with multiple attractors were used to optimize the system.The optimal parameters of the multiple attractor cellular automata classifier were improved through genetic algorithm, thus improving the big data classification method.Experimental results show that this method can effectively identify and classify structured data and unstructured data in non relational distributed submission information, and has high classification accuracy.

Research on the effectiveness evaluation of cyber security curriculum construction for combating telecom network fraud

Shengli ZHOU, Keyi JIANG, Bo XU, Rui XU, Xikang ZHANG, Quanzhe ZHAO, Yangdong XU

2023, 39(6):  122-128.  doi:10.11959/j.issn.1000-0801.2023127

Asbtract ( 123 )   HTML ( 20)   PDF (1752KB) ( 166 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The governance of telecom network fraud under the new situation puts forward new requirements for developing network security courses of public security colleges.Considering the cyber security curriculums in public security colleges and universities that were not adaptive or practical enough in terms of training objectives, training plans, evaluation systems, etc, the methods such as chi-square test and independent samples t-test were used to analyze the construction status of cyber security curriculums in major public security universities and colleges nationwide, and the quantitative evaluation of course effectiveness was achieved by combining score card mathematical analysis.Based on the data analysis, an optimization path for the construction of cyber security curriculums in major public security universities and colleges was proposed.

Communication Standard

Research on 6G standardization of International Telecommunications Union（ITU）

Yan WANG, Ying PENG

2023, 39(6):  129-138.  doi:10.11959/j.issn.1000-0801.2023130

Asbtract ( 285 )   HTML ( 67)   PDF (1219KB) ( 445 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The International Telecommunications Union (ITU), as the authoritative international standardization organization in the field of information and communications (ICT), is leading the pace of the work in technology research and standardization of 6G.With the joint efforts of various parties in the global industry, it expected to officially release international standards for 6G around 2030 and promote its commercialization.Firstly, the basic information and organizational structure of ITU was introduced.And the current progress of ITU’s IMT-2030(6G) standardization work including the Recommendation “ITU-R M.[IMT.Framework for 2030 and Beyond]”, Report “Future technology trends of terrestrial IMT systems towards 2030 and beyond” and etc., as well as subsequent standardization research were analyzed.In addition, the impact of ITU’s 6G standardization work on the mobile communication and industries was also analyzed.The standardization work of 6G in ITU can not only promote technological changes and upgrades for the new generation of mobile communication systems, but also drive the transformation and development of the industry ecosystem.

Engineering and Application

OSU based M-OTN networking modes and field trial

Ruiquan JING, Leiyang SHEN, Zhiwei JIN, Jing ZHANG, Yuan YAO, Lei CHENG, Heng ZHOU

2023, 39(6):  139-148.  doi:10.11959/j.issn.1000-0801.2023124

Asbtract ( 137 )   HTML ( 32)   PDF (3149KB) ( 137 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The innovative optical service unit (OSU) based on M-OTN technologies can provide high-quality private line services with low latency and flexible bandwidth, and improve the networking capability and efficiency of OTN for small granularity services.The application mode of M-OTN/OSU technologies in the OTN were discussed, including the value OSU brings, the comparison of Overlay vs.Underlay networking modes, multiplexing modes of OSU into ODUk and multi-service unified transport.The service capability and long-term operational reliability of OSU technology were fully verified through the field trial network, which provides a technical basis for the selection of OSU networking modes.

Research on digital twin network capability maturity model of telecom operators

Shanshan LI, Jianhong XU, Tao LI, Yongjian ZHAO, Zhifei LIU

2023, 39(6):  149-158.  doi:10.11959/j.issn.1000-0801.2023118

Asbtract ( 198 )   HTML ( 43)   PDF (1803KB) ( 381 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The digital twin network is an important direction for the transformation and upgrading of the communication industry, and a key lever for telecom operators to implement national strategies and empower the digital economy.The research results of digital twin maturity in the industry were drawn on, the connotation of digital twin network was analyzed, and a set of digital twin network capability maturity model and evaluation method suitable for telecom operators was put forward.Through systematic description of the development stage of digital twin networks and quantitative evaluation of the maturity of digital twin networks capability, it is helpful for operators to identify technical or organizational weaknesses, improve expected results, and promote innovation and collaboration among different organizations in the industry.

Research and application of 5G lightweight hardware encryption module for power terminals

Chen ZHANG, Hongkai WANG, Dong MAO, Sichen PAN, Shuai ZHAO

2023, 39(6):  159-169.  doi:10.11959/j.issn.1000-0801.2023104

Asbtract ( 103 )   HTML ( 17)   PDF (2878KB) ( 138 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

With the accelerated deployment of 5G new infrastructure, terminal application business has been gradually extended to various vertical industries.Although 5G technology can also adapt to the differentiated communication requirement for power industry, the power industry currently lacks a kind of 5G module that not only meets the security access requirement for power grid, but also has the advantage of low hardware upgrade cost, flexibility and scalability.Therefore, a 5G lightweight hardware encryption module for power terminals was designed and developed.The module is connected with the power terminal in the form of pluggable interface, and the capability of 5G communication and encryption authentication could be quickly realized just by terminal software upgrade.The test results show that the module has chip-level security authentication capability, and can take into account 5G communication and security mechanism, which not only improves the security of terminal access to the power intranet, but also reduces the difficulty of terminal transformation and 5G communication hardware upgrade cost.