Please wait a minute...

Current Issue

    20 October 2021, Volume 37 Issue 10
    Comprehensive Review
    Applications and challenges of silicon-based optoelectronics in communication
    Dezhao LI, Pengfei XU, Kejian ZHU, Zhiping ZHOU
    2021, 37(10):  1-11.  doi:10.11959/j.issn.1000-0801.2021247
    Asbtract ( 430 )   HTML ( 57)   PDF (888KB) ( 692 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Silicon-based optoelectronics technology has the advantages of photonics, such as broadband, high-speed and anti-interference, as well as the advantages of microelectronics, such as large-scale integration, low energy consumption and low cost.Silicon-based optoelectronics is showing superior advantages in high-speed and short-reach communications and posing a challenge to the conventional indium phosphide (InP) optical modules.The application, development and technical challenges of silicon-based optoelectronics technology in communications were briefly discussed.There is an urgent need to develop high performance devices, packaging technology, automation software for further development of the silicon-based optoelectronic technology.

    Network intelligence standards, open source and industry research
    Lingli DENG, Ninglun GU, Xiangyang YUAN, Li YU, Junlan FENG, Chao DENG, Zhiqiang YANG, Tongxu ZHANG, Rongyu WU, Xinpeng ZHANG
    2021, 37(10):  12-21.  doi:10.11959/j.issn.1000-0801.2021217
    Asbtract ( 334 )   HTML ( 40)   PDF (1078KB) ( 417 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Network intelligence is the deep integration of AI technology and communication network hardware, software, systems, processes, etc.to improve process intelligence, reduce costs, increase efficiency, improve quality, accelerate technology innovation, and enable service agility.The autonomous network proposes to optimize user experience, automate management operations, and maximize resource efficiency by simplifying the network architecture, encapsulating autonomous domains, and providing closed loops for business/network operation control, which clarifies the target architecture and implementation path for network intelligence.Firstly, based on the layered architecture and evaluation framework of the autonomous network, the network intelligent technology system was summerized.Secondly, extensive research on relevant standards organizations, open source communities, industrial collaboration, and R&D and application status were conducted.Finally, gap analysis, collaborative suggestions and summary prospects for guiding the following-up industry development were provided in terms of the application requirements from operators in related practices.

    Topic: Data Network and Computing Power Network
    Internet design concepts and network 5.0 technologies
    Lintao JIANG, Xiuying NIE, Jie ZHANG
    2021, 37(10):  24-30.  doi:10.11959/j.issn.1000-0801.2021232
    Asbtract ( 212 )   HTML ( 23)   PDF (544KB) ( 155 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The internet has been developed for more than 50 years, but its design concept has never changed.The business network is completely OTT, IP network is still “best effort”, which restricts the development of the internet.In order to continuously meet the needs of social development, the internet design idea needs to be changed, and the feasible direction is to be from “application-centered” to the “network-centered”.Network 5.0 is the fifth generation of data network, compared with the fourth generation of data network (IP network), it retains all the advantages of IP network, filling the absence of IP network in network performance and ability, with qualitative progress in performance and ability.The key technical issues on scalability, endogenous credible security, certainty and more aspects of network 5.0 were specifically discussed.

    Full-dimensional definable network capabilities
    Congshan FAN, Xu ZHOU, Yifang QIN, Taixin LI
    2021, 37(10):  31-38.  doi:10.11959/j.issn.1000-0801.2021235
    Asbtract ( 174 )   HTML ( 12)   PDF (769KB) ( 106 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    To solve the problems of single form of existing network capabilities, poor dynamics, low efficiency, and difficulty in adapting to the needs of specialized and personalized services in complex scenarios in the future, a full-dimensional definable network capability model was proposed.Based on the design concept of “network centric”, each layer of the network is opened up to realize the definition of full-dimensional network capabilities such as communication subjects, network functions, network resources, and security, and supports dynamic loading and evolutionary development.Through self-mapping of service requirements and self-organization of network capabilities, it can flexibly adapt to service development needs, and actively output diversified network endogenous capabilities to terminal services in a variety of interface forms, which will naturally adapt to the complex and uncertain evolution of future services.

    Future IP network architecture evolution for service and network collaboration
    Bing HUANG, Bin TAN, Jian LUO, Yong GUO
    2021, 37(10):  39-46.  doi:10.11959/j.issn.1000-0801.2021234
    Asbtract ( 174 )   HTML ( 25)   PDF (1571KB) ( 231 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    As the technical foundation of the internet, how IP technology develops in the next 5 to 10 years is the key issue facing the communication industry.The success and weakness of the traditional IP network architecture were analyzed, and the future IP network should not only inherit the design principles of the traditional IP technology was proposed, such as the end-to-end principle and layered decoupling principle, but also solve the problem that the service is too separated from the network.From the perspective of network evolution and future service requirements, the core issue of future IP networks is to strengthen the collaboration of between services and networks , and the improved IP network design principle “horizontal: end-to-end principle empowered by service-oriented network, vertical: layered model supported by intelligent control plane” was proposed, and based on this principle, the reference architecture of the future network was proposed.

    Ubiquitous IP protocol system
    Xu ZHOU, Sheng JIANG, Jie ZHANG, Bo LEI, Xiuli ZHENG, Taixin LI
    2021, 37(10):  47-54.  doi:10.11959/j.issn.1000-0801.2021237
    Asbtract ( 269 )   HTML ( 16)   PDF (730KB) ( 149 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Considering the inherent characteristics of the TCP/IP protocol stack are difficult to carry the personalized needs of the future network, based on the development of the network protocol, the next-generation ubiquitous IP (UIP) protocol system was proposed.Ubiquitous IP is a complete set of new network technology system based on IP protocol expansion and “progressive” and “revolutionary” innovation.The key content includes the design concept with flexible IP as the core, the three-dimensional protocol system, and the protocol innovation in the network layer and transport layer IPv6+ and UIP, as two-generation network innovation technologies, can relay the improvement of network capabilities were compared and analyzed.The current progress and future prospect of UIP were summarized.

    Global cyclic queuing and forwarding mechanism for large-scale deterministic networks
    Yijun MO, Zihan YANG, Huiyu LIU, Tianliu HE
    2021, 37(10):  55-65.  doi:10.11959/j.issn.1000-0801.2021240
    Asbtract ( 285 )   HTML ( 37)   PDF (852KB) ( 379 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Delay and jitter sensitive industrial control and remote driving application-driven network by best-effort to deterministic service.To meet the deterministic application of delay and jitter demands, industry and academia have conducted detailed research on stream filtering shaping and queue forwarding control, implementing microsecond delay and microsecond jitter of light load steady in small-scale ethernet environment.However, the study seldom considers the effects of long-distance topology change, transmission delay fluctuations and short-time traffic overloads, challenging to meet deterministic requirements under large-scale backbone networks.For this challenge, based on cyclic queuing and forwarding, considers source-destination path properties and deterministic traffic intensity, the global cyclic queuing and forwarding three-queue was proposed.The deterministic network simulation system was based on OMNeT++.Experiments compare queue forwarding control mechanisms such as CQF, CQF-three and GCQF-three.Results show that GCQF-three can timely adjust the gatecontrol and queueing priority according to the deterministic traffic state, achieves the best in delay, jitter and network utilization.

    Klonet: a network emulation platform for the technology innovation
    Jingzhao XIE, Wei SHAN, Chang XIAO, Tie MA, Li CHEN, Hongfang YU, Gang SUN
    2021, 37(10):  66-75.  doi:10.11959/j.issn.1000-0801.2021241
    Asbtract ( 499 )   HTML ( 35)   PDF (808KB) ( 246 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The innovations of key technologies enable and promote the future development of networks and the network industry.Therefore, there is an urgent need for a network experimental platform that can easily and quickly verify innovative network technologies, so asto effectively lower the innovation barriers and promote innovations.By analyzing the advantages of network emulation technology in network experiments and combining the demand of network technology innovation for network experiment platform, a network emulation experiment platform Klonet for network technology innovation was proposed.The platform can flexibly expand the scale of the network, emulate the network in all perspectives, manage the emulated network in a fine-grained manner, and thus ease the workflow of network experimentation.The experimental method on Klonet was described and the ability of Klonet to support network technology innovation was verified based on two specific use cases.

    Computing force network technologies for computing and network integration evolution
    Xiaodong DUAN, Huijuan YAO, Yuexia FU, Lu LU, Tao SUN
    2021, 37(10):  76-85.  doi:10.11959/j.issn.1000-0801.2021248
    Asbtract ( 1487 )   HTML ( 220)   PDF (855KB) ( 1554 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the construction of the national hub node of the integrated computing force network and the accelerated implementation of the “eastern data and western calculation” project, the integration of computing and network is moving towards the deep water area. Facing the needs of emerging businesses for networks and computing, how to coordinate distributed computing resources, service applications and network resources to meet service needs has become an urgent problem to be solved. Aiming to solve the problems of computing and network awareness, coordination, and scheduling. The key technologies, including phased evolution of computing force network , computing-aware technology architecture, computing force quantity and identification, computing force routing and other technologies were explored and researched, and a variety of deployment modes of computing force network was proposed, and reference was provided for the follow-up technology research and industry development of the computing force network.

    A solution of computing power network based on domain name resolution
    Qianying ZHAO, Wenjuan XING, Bo LEI, Lintao Jiang
    2021, 37(10):  86-92.  doi:10.11959/j.issn.1000-0801.2021233
    Asbtract ( 297 )   HTML ( 32)   PDF (1243KB) ( 337 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the development of new digital technologies and businesses, and the explosive growth of information and data, cloud - edge - end multilevel computing power resources are facing huge application requirements.Generalization of computing power infrastructure has become a major trend.As a new direction of network technology, computing power network integrates computing and other resources with the network, provides optimal resource configuration policies for users based on their requirements, and improves the cooperative efficiency of multilevel computing resources.The current technical route of the computing power network were analyzed, a solution of computing power network based on domain name resolution was proposed.The solution introduces a domain name resolution mechanism to use URL language to identify multiple computing power resources uniformly, and a centralized computing power resource management platform performs unified allocation and scheduling of computing power resources.After the user receives the assigned computing power resource identifier, the network information of corresponding computing resource were parsed through the domain name resolution system, and the network connection was established with the resource pool through the computing power gateway for the user to access.This solution meets the demand for flexible expansion of the computing power resource identification and has good practicability and universality.

    Some key technical issues in the evolution from cloud network to computing network
    Chang CAO, Shuai ZHANG, Ying LIU, Xiongyan TANG
    2021, 37(10):  93-101.  doi:10.11959/j.issn.1000-0801.2021231
    Asbtract ( 268 )   HTML ( 35)   PDF (885KB) ( 388 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Firstly, the development trend of cloud-network integration technology in recent years, the new problems and challenges it faces were analyzed.Then, based on the heterogeneous computing power, the sinking of computing power, and the new changes in computing-network linkage in business development, six major computing and network capabilities were proposed.Convergence requirements, combined with multi-cloud connection, differentiated bearer, and hierarchical security, discussed the case of building new capabilities and new business experience around the computing network.The analysis results show that it can be concluded that the computing network and the cloud network are closely related in the technical system, and the integration of the computing network is the inevitable result of the development and evolution of the integration of the cloud and the network, and it is also an inevitable choice for the development of the digital economy.

    Research and Development
    5G urban rail traffic scenario classification and channel modeling
    Ruisi HE, Bo AI, Zhangdui ZHONG, Mi YANG, Chen HUANG, Zhangfeng MA, Guiqi Sun, Hang MI, Chengyi ZHOU, Ruifeng CHEN
    2021, 37(10):  102-116.  doi:10.11959/j.issn.1000-0801.2021243
    Asbtract ( 340 )   HTML ( 23)   PDF (1573KB) ( 388 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Urban rail traffic is an important part of modern transportation infrastructure.As a new generation of mobile communication technology, 5G can provide high data rate and low latency wireless transmission, which helps to improve the efficiency and service quality of urban rail traffic system.Due to the complexity of urban rail traffic scenarios, accurate communication scenario classification, channel characterization and channel models are required to provide theoretical support for the design of urban rail traffic 5G communication systems.The classification of 5G urban rail traffic radio propagation scenarios to support channel measurements and modeling was proposed.Current status of urban rail traffic channel measurements and modeling was shown, and the current challenges were analyzed.The applications of artificial intelligence in channel feature extraction and channel modeling were discussed, and the 5G urban rail traffic channels by considering reconfigurable intelligent surface and unmanned aerial vehicle were analyzed.Finally, the research on 5G urban rail traffic channel modeling at millimeter wave frequency band was described.

    Application of blockchain in urban rail traffic edge computing network
    Gaochang XIE, Hua LU, Qinqin TANG, Han ZHU, Chenghao LIANG, Wen WEN, Renchao XIE
    2021, 37(10):  117-125.  doi:10.11959/j.issn.1000-0801.2021238
    Asbtract ( 261 )   HTML ( 29)   PDF (832KB) ( 459 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Multi-access edge computing (MEC) can provide high-quality service capabilities for computing-intensive services and delay-sensitive services in urban rail traffic.However, many edge facilities in rail traffic edge computing network are exposed to an open environment, and their privacy protection and transmission security are facing great challenges.Blockchain has functional characteristics such as distributed ledger, consensus mechanism, smart contract, and decentralized application.Therefore, the use of blockchain can build a systematic security protection mechanism for the distributed rail traffic edge computing network to ensure network security and data security and realize high-quality urban rail traffic services.Firstly, the basic concept of the blockchain and the urban rail traffic edge computing network architecture were introduced.Then, the structure and application content of the rail traffic edge computing network security protection mechanism integrated with the blockchain was discussed in detail.Finally, the open research issues and challenges of the security protection mechanism were analyzed.

    Parameter estimation of frequency hopping signal based on time-frequency matrix LCM
    Jiamin LIU, Zhijin ZHAO, Junna SHANG, Xueyi YE
    2021, 37(10):  126-135.  doi:10.11959/j.issn.1000-0801.2021239
    Asbtract ( 122 )   HTML ( 6)   PDF (842KB) ( 78 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    To improve the performance of the parameter estimation of frequency hopping signal under low signal-to-noise ratio, a parameter estimation algorithm of frequency hopping signal based on time-frequency matrix LCM was proposed.Considering the different characteristics of time-frequency distribution of frequency hopping signal and noise, the mean local energy comparison value of the time-frequency matrix under different scale sliding windows was used, and the multi-scale local energy comparison feature matrix was introduced, so the time-frequency matrix P that only contains the time-frequency information of the frequency hopping signal could be obtained through this feature matrix and the adaptive threshold separation.And then, by extracting the time-frequency hopping information from P, the period, hopping time and frequency of the frequency hopping signal could be accurately estimated.The simulation results show that compared with the traditional local contrast measure (LCM) and morphological filtering method, the proposed algorithm has better frequency hopping signal extraction effect and higher parameter estimation accuracy.And then the tests of the algorithm on the DSP+FPGA hardware system verify the effectiveness and practicability.

    Application of improved self-training model in the identification of users with poor service quality
    Li YU, Zhe LI, Fei GAO, Xiangyang YUAN, Yong YANG
    2021, 37(10):  136-142.  doi:10.11959/j.issn.1000-0801.2021191
    Asbtract ( 164 )   HTML ( 37)   PDF (593KB) ( 109 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Poor quality user identification is an important method to reduce the complaint rate and increase satisfaction.It is difficult to effectively label a large amount of structured and unstructured data related to business perception in current telecommunications network systems, poor quality user labels are not complete, and the existing supervised learning model training samples are unbalanced, resulting in a low quality recognition rate.An improved self-training semi-supervised learning model was adopted, a small number of low-satisfaction and complaint users as poor quality user labels was used to label network data, and label migration was used to train a large amount of unlabeled data to identify poor quality users.Experiments show that compared to fully supervised learning with high recognition model accuracy but high training cost and unsupervised learning with low recognition model accuracy, semi-supervised learning can make full use of unlabeled sample data for effective training, ensuring lower training costs and the recognition accuracy of poor-quality users is significantly improved.

    Engineering and Application
    Networking technology and implementation method based on 5G slice private line
    Yan ZHANG
    2021, 37(10):  143-151.  doi:10.11959/j.issn.1000-0801.2021230
    Asbtract ( 329 )   HTML ( 42)   PDF (3550KB) ( 150 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the progressive application of 5G technology in 2B user network, the integration of fixed and mobile collaborative networking and DICT technology is the trend of future development.The establishment of fixed line from 5G network wireless access to 2B user site or cloud system, was analyzed the STN line, OTN line of fixed network, and the 5G network slice defined by 3GPP were used to study and compare the networking characteristics of 5G-STN, 5G-STN into cloud and 5G-OTN of 5G slice line, and the end-to-end specific network capabilities.The fixed-mobile collaborative networking scheme was tested on the operator's network to consolidate and verify 5G slicing private line networking technology, drive the implementation of 5G slicing private line networking technology, and accelerate the integration of “5G+” and the real economy.

    Orchestration system architecture design and implementation of smart line service
    Huiguang CHEN, Huan LI
    2021, 37(10):  152-161.  doi:10.11959/j.issn.1000-0801.2021209
    Asbtract ( 203 )   HTML ( 19)   PDF (1064KB) ( 122 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In the era of SDN/NFV technology, in order to meet the goal of rapid and automatic activation of new network services, a solution about the architecture and implementation of orchestrator system was introduced.According to the characteristics of loose coupling and flexible deployment of each function, the orchestrator’s system architecture was divided into service entry, network and adapter.The choreographer workflow was divided into the business preparation workflow and the business delivery workflow.Through practicing and verifying, as the core system, the smart line basically met the initial business requirements and maintenance management requirements of the business choreographer.

    Privatize modification of broadband dual IP stack on cloud platform
    Chengli ZHANG, Xiaorong HE, Weifang SONG
    2021, 37(10):  162-170.  doi:10.11959/j.issn.1000-0801.2021115
    Asbtract ( 130 )   HTML ( 12)   PDF (798KB) ( 116 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    As IPv4 address became exhausted even sooner, China Telecom had been speeding up the IPv4/IPv6 dual IP stack construction before the final large scale application of IPv6.Meanwhile, emphasis of IPv4 address privatize modification was placed on following schedule at this very moment to ensure the smooth transition and public IPv4 address providing.By studying the trace-back model of IPv6 and private IPv4 dual stack IP address in typical telecom business scenarios, the way to separate IP address trace-back model from traditional broadband AAA platform and migrated into cloud deployment were acquired , thus provided a reference for the AAA entirety reconstruction.

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: