Current Issue

20 March 2021, Volume 37 Issue 3

Previous Issue    Next Issue

Comprehensive Review

Next decade of telecommunications artificial intelligence

Ye OUYANG, Lilei WANG, Aidong YANG, Maulik SHAH, David BELANGER, Tongqing GAO, Leping WEI, Yaqin ZHANG

2021, 37(3):  1-36.  doi:10.11959/j.issn.1000-0801.2021055

Asbtract ( 2490 )   HTML ( 433)   PDF (3240KB) ( 2173 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

It has been an exciting journey since the mobile communications and AI were conceived 37 years and 64 years ago.While both fields evolved independently and profoundly changed communications and computing industries, the rapid convergence of 5G and AI is beginning to significantly transform the core communication infrastructure, network management and vertical applications.The individual roadmaps of mobile communications and artificial intelligence in the early stage were firstly outlined, with a concentration to review the era from 3G to 5G when AI and mobile communications started to converge.With regard to telecommunications artificial intelligence, the progress of AI in the ecosystem of mobile communications was further introduced in detail, including network infrastructure, network operation and management, business operation and management, intelligent applications towards BSS ＆amp; OSS convergence, verticals and private networks etc.Then the classifications of AI in telecom ecosystems were summarized along with its evolution paths specified by various international telecommunications standardization bodies.Towards the next decade, the prospective roadmap of telecommunications artificial intelligence was forecasted.In line with 3GPP and ITU-R’s timeline of 5G ＆amp; 6G, the network intelligence following 3GPP and O-RAN routes, experience and intention driven network management and operation, network AI signaling system, intelligent middle-office based BSS, intelligent customer experience management and policy control driven by BSS＆amp; OSS convergence, evolution from SLA to ELA, and intelligent private network for verticals were further explored.It concludes that with the vision AI will reshape the future B5G/6G landscape, and we need fully take the unprecedented opportunities.

Topic: Endogenous Safety and Security

Key technologies and implementation methods of endogenous safety and security cloud data center based on mimic architecture

Fan ZHANG, Guangwei XIE, Wei GUO, Hongchao HU, Ruyun ZHANG, Wenyan LIU

2021, 37(3):  39-48.  doi:10.11959/j.issn.1000-0801.2021056

Asbtract ( 246 )   HTML ( 27)   PDF (1077KB) ( 400 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Cloud data center is the representative of the new generation of information infrastructure, and its security has become the focus of attention in recent years, which is of great significance.Based on the analysis of the current cloud security situation, the security architecture, key technologies, and implementation methods of the cloud data center through the emerging concept of endogenous security were explored, hoping to use the mimic structure to solve the endogenous safety and security problems such as vulnerabilities and backdoors that were difficult to deal with by the existing means.Moreover, the endogenous security architecture of cloud data centers and related key technologies were proposed, with the mode and trend of mimic transformation.In the future, endogenous safety and security cloud data centers will provide practical solutions for the construction of a new generation of information infrastructure, which may accelerate the technology application and promotion of the cloud service model.

An endogenous safety and security technical framework inspired by biological immune system

Yufeng LI, Chenhong CAO, Jiangtao LI, Peng WANG

2021, 37(3):  49-56.  doi:10.11959/j.issn.1000-0801.2021051

Asbtract ( 241 )   HTML ( 25)   PDF (783KB) ( 424 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Inspired by the endogenous safety and security effect of biological immune mechanism, an endogenous safety and security technology framework was proposed, the main technical challenges of the framework were analyzed, the corresponding solutions were given, and the application scenarios of the framework were analyzed.Comprehensive analysis shows that the framework presents high robustness, autonomy and adaptability, and has strong practicability.

Mimic security scheduling algorithm for industrial control under limited heterogeneous resource constraints

Ruyun ZHANG, Heyuan LI, Shunbin LI

2021, 37(3):  57-65.  doi:10.11959/j.issn.1000-0801.2021047

Asbtract ( 143 )   HTML ( 16)   PDF (747KB) ( 291 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Cyberspace mimic defense technology is an effective method to deal with backdoor attacks on unknown vulnerabilities in information systems.Its security is closely related to the number and the heterogeneity of the executors and the scheduling strategy.However, in the field of industrial control, the ecological resources of industrial application are relatively closed, and the number of realizable heterogeneous executors is limited.To solve the above problems, a mimic scheduling algorithm for industrial control under the constraints of limited heterogeneous resources was proposed.The experimental results show that the proposed algorithm for triple-redundancy mimic industrial control system was able to select a suitable executor to go online according to environmental characteristics adaptively.Even in a high-intensity attack environment, it can still maintain a high availability probability of 99.24%.

Management technologies of cyberspace endogenous safety and security test site

Hongyi ZHU, Xiaoyuan LU, Yi LI

2021, 37(3):  66-74.  doi:10.11959/j.issn.1000-0801.2021048

Asbtract ( 266 )   HTML ( 20)   PDF (756KB) ( 444 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The cyber range has been widely recognized as an effective way to study the technologies of network attack/defense and the vulnerability of network architecture.The endogenous safety and security test site that receives high attention recently is a type of cyber range for cyber space endogenous safety and security technologies.Based on the new situation of cyberspace with 5G development, a test site management technology for virtual-real network settings was proposed, and a test site management architecture based on the endogenously secured software defined network controller was designed.A design of the endogenously secured network control system was also proposed, which used a middle layer forwarding agent to achieve data isolation and supported multiple heterogeneous open-source controllers.Finally, a test site scenario reconfiguration and resource orchestration method was proposed based on a test site networking scheme.

Resource public key infrastructure RPKI: status and problems

Yingying SU, Dan LI, Honglin YE

2021, 37(3):  75-89.  doi:10.11959/j.issn.1000-0801.2021050

Asbtract ( 513 )   HTML ( 81)   PDF (870KB) ( 592 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The BGP (border gateway protocol) did not fully consider security issues at the beginning of its design.With the rapid growth of the Internet, its security risks have become incrementally obvious.Academia and industry have proposed many solutions to the security issues faced by inter-domain routing, and what is really deployed is the RPKI (resource public key infrastructure) promoted by the IETF (the Internet Engineering Task Force).The development status and research advances of RPKI were surveyed, with emphasis on problems of RPKI, existing solutions, and related limitations.Moreover, latest achievements on RPKI function extension were introduced.Finally, future research directions were concluded.

Cloud data anonymous assured deletion approach based on blockchain

Shuangxing WANG, Jintang LUO, Lisha SHUAI, Jiamin ZHANG, Min ZHANG, Xiaolong YANG

2021, 37(3):  90-104.  doi:10.11959/j.issn.1000-0801.2021049

Asbtract ( 215 )   HTML ( 17)   PDF (1141KB) ( 431 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The previous assured deletion schemes ignored that the privacy is threated by the relevance between data and user identity, exposing user behavior to the cloud service provider or attackers.Aiming at the above problems, an assured deletion approach based on blockchain was proposed.The improved linkable ring signature scheme was adopted to make user achieve anonymous assured deletion by setting specific link symbol in the signature.Moreover, the blockchain technology was leveraged to guarantee the nonrepudiation of assured deletion.Theoretical analysis and experimental results show that the proposed approach can meet the assured deletion requirements of user data in cloud storage, and its anonymity can effectively prevent attackers from tracking and analyzing user behavior.

Research on network traffic classification based on machine learning and deep learning

Yue GU, Dan LI, Kaihui GAO

2021, 37(3):  105-113.  doi:10.11959/j.issn.1000-0801.2021052

Asbtract ( 595 )   HTML ( 85)   PDF (698KB) ( 635 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

With the continuous development of Internet technology and the continuous expansion of network scale, there are many different types of applications , and various new applications have endlessly emerged.In order to ensure the quality of service (QoS) and ensure network security, accurate and fast traffic classification is an urgent problem for both operators and network managers.Firstly, the problem definition and performance metrics of network traffic classification were given.Then, the traffic classification methods based on machine learning and deep learning were introduced respectively, the advantages and disadvantages of these methods were analyzed, and the existing problems were expounded.Next, the related work by focusing on the three problems encountered elaborated and analyzed in traffic classification when considering online deployment: dataset, zero-day application identification and the cost of online deployment, and further discusses the challenges faced by the current network traffic classification researches.Finally, the next research direction of network traffic classification was prospected.

Research and Development

A reduced-complexity compressed sensing channel estimation for underwater acoustic channel

Xuan YU, Xuan GENG

2021, 37(3):  114-124.  doi:10.11959/j.issn.1000-0801.2021039

Asbtract ( 145 )   HTML ( 12)   PDF (943KB) ( 319 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the sparse characteristics of underwater acoustic channels for shallow seas, a reduced-complexity look-ahead backtracking orthogonal matching pursuit (RC-LABOMP) channel estimation algorithm was proposed.Firstly, two types of support sets of orthogonal matching pursuit and subspace pursuit channel estimation algorithms were calculated, and then prior information based on the intersection and union of the two support sets were preprocessed.At last, the preprocessed prior information was used to complete look-ahead backtracking orthogonal matching pursuit channel estimation.The preprocessed prior information leads to the decrease of the iteration number of original LABOMP, and reduction of the atom index range, thus the proposed algorithm can reduce the computational complexity of original LABOMP significantly.In addition, combining the proposed algorithm with the underwater acoustic Turbo equalization system is more suitable for underwater acoustic communication systems.Simulation results show that the proposed algorithm demonstrates high estimation accuracy and low bit error rate performance under both conditions of random channels and underwater acoustic channels.It also reduces the computational complexity of the LABOMP algorithm.Therefore, it is an effective method for shallow seas underwater acoustic channels estimation algorithm.

A novel method for encoding and decoding based on enhanced parity-check-concatenated polar codes

Yan WANG, Shunlan LIU, Zhenzhen XI, Jianrong BAO

2021, 37(3):  125-132.  doi:10.11959/j.issn.1000-0801.2021033

Asbtract ( 211 )   HTML ( 10)   PDF (866KB) ( 360 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Polar codes have perfect coding and decoding performance as a kind of error correction code, which have become a standard coding scheme for 5G short code control channel. While the length of polar codes is short, its performance is not good enough. A novel method for encoding and decoding based on enhanced parity-check-concatenated polar codes was proposed. The enhanced check bit was set up after the parity-check (PC) bit to check the information bits with low channel reliability again, which could help path metric value prune the decoding path and improve the reliability of decoding path selection. The simulations reveal that the proposed method has better performance than cyclic redundancy check (CRC) concatenated polar codes and PC concatenated polar codes under the same channel, code length and code rate conditions. In additive white Gaussian noise channel, when the code length is 128, the code rate is 1/2 and the bit error rate is 10^-3, the proposed method achieves a gain of about 0.3 dB over the PC codes cascaded polar codes, about 0.4 dB over the CRC assisted polar codes.

Vibe++ background segmentation method combining MeanShift clustering analysis and convolutional neural network

Zihao LIU, Xiaojun JIA, Sulan ZHANG, Zhiling XU, Jun ZHANG

2021, 37(3):  133-145.  doi:10.11959/j.issn.1000-0801.2021046

Asbtract ( 223 )   HTML ( 13)   PDF (2228KB) ( 179 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve problems of noise points and high segmentation error for image shadow brought by traditional Vibe+ algorithm, a novel background segmentation method (Vibe++) based on the improved Vibe+ was proposed.Firstly, binarization image was acquired by using traditional Vibe+ algorithm from surveillance video.The connected regions were marked based on the region-growing domain marker method.The area threshold was obtained with difference characteristics of boundary area, the connected regions below threshold were treated as disturbing points.Secondly, five different kernel functions were introduced to improve the traditional MeanShift clustering algorithm.After improving, this algorithm was fused effectively with partitioned convolutional neural network.Finally, program of classification of trailing area, non-trailing area and trailing edge area in the resulting image was performed.Position coordinates of the trailing area were calculated and confirmed, and the trailing area was quickly deleted to obtain the final segmentation result.This segmentation accuracy was greatly improved by using the proposed method.The experimental results show that the proposed algorithm can achieve segmentation accuracy of more than 98% and has good application effect and high practical value.

RSS localization algorithm based on second-order cone programming in non-line-of-sight environment

Xiaoping JIN, Jun LIANG, Shaofeng XIE

2021, 37(3):  146-153.  doi:10.11959/j.issn.1000-0801.2021032

Asbtract ( 140 )   HTML ( 8)   PDF (856KB) ( 187 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the problem that the accuracy of localization using received signal strength (RSS) in wireless sensor network is insufficient in non-line-of-sight (NLOS) environment, a new robust localization algorithm based on second-order cone programming (SOCP) was proposed.Based on the assumption of the upper bound of NLOS biases, a location equation with robustness to NLOS biases was constructed to suppress the interference of NLOS biases, then convex optimization was used to transform the robust localization problem into a second-order cone programming problem with the intention of accurate estimation, thereby improved the localization accuracy.In addition, the localization problem was extended to the case of unknown transmission power, and an iterative SOCP algorithm was proposed.Simulation results show that the proposed algorithm can effectively solve the problems existing in NLOS localization, and the localization accuracy is better than Newton iteration method, UT method and SOCP method.

Engineering and Application

Trend of mobile communication market pattern under the reform of industry co-competition

Su DIAO, Cheng JIANG

2021, 37(3):  154-161.  doi:10.11959/j.issn.1000-0801.2021053

Asbtract ( 149 )   HTML ( 17)   PDF (999KB) ( 311 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In October 2019, with the nationwide implementation of mobile number portability, MIIT and SASAC strengthened industry competition supervision and initiated the industry co-competition of the three major telecom operators in China.The short-term development trend of China mobile communications market was predicted by utilizing Markov chain model to analyze the changes in the market share of mobile communication market, and the state transition probability matrixes before and after the change of the industry’s competitive environment caused by regulatory policies were innovatively compared.The study shows that the impact of industry co-competition reform is greater the nationwide implementation of MNP, the probability of mobile users to lock-in has increased.According to the prediction of the state transition probability matrix after industry co-competition, China Telecom’s mobile market share will increase, while China Mobile and China Unicom Mobile’s market share will decline in the short and medium term.However, China Mobile will continue to maintain its dominant position.To accelerate 5G high-quality development and open up a new pattern in the mobile communication market, the industry co-competition environment needs to be further optimized to strengthen regulatory policy, 5Gco-construction, and operation management.

Network quality and user perception improvement of high-speed rail scene

Songbai LIANG, Yingwu GUO, Xinwei LI

2021, 37(3):  162-170.  doi:10.11959/j.issn.1000-0801.2021045

Asbtract ( 202 )   HTML ( 27)   PDF (1435KB) ( 284 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In view of the difficulty in improving the quality of mobile network in high-speed railway scenes and the poor efficiency of conventional methods, a methodology system for improving the perception of mobile communication users was explored in traffic trunk lines, especially in high-speed railway scenes.The methodology system was based on handover chain optimization with the core idea of “primary covering, reasonable reselection and orderly switching”.It adopted a three-layer theoretical structure include covering control firstly, then parameter optimization, and feature loading last.It used idle state and service state data pull graph correlation to determine problem road sections and cells, match the delay parameters of high-speed mobile scenes, load service quality and perception characteristics, and greatly improved the high-speed railway network quality and user perception.This method started from the principle analysis, with practical application, it could effectively solve the network problems, which could be fully extended to other linear scenes.