Please wait a minute...

Current Issue

    20 February 2022, Volume 38 Issue 2
    Reviews
    Research progress and trends of deep learning based wireless communication receiving method
    Panpan LI, Zhengxia XIE, Guangxue YUE, Xin LIU
    2022, 38(2):  1-17.  doi:10.11959/j.issn.1000-0801.2022025
    Asbtract ( 519 )   HTML ( 103)   PDF (675KB) ( 652 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the continues expansion of the application boundary for wireless communications, the application environment of wireless communications is becoming increasingly complex and diverse, which faces negative impacts such as radio frequency (RF) damage, channel fading, interference and noise.It brings difficulties to recover the original information at the receiver.Drawing from the research results of deep learning methods in computer vision, pattern recognition, natural language processing and other fields, wireless communication reception technology based on deep learning has received wide attentions from both academia and industry.Firstly, the current research status of wireless communication reception technology based on deep learning at home and abroad was described.Secondly, the current technical challenges of wireless communication reception in the context of signal big data were outlined, and a reference architecture of intelligent wireless communication reception based on deep neural network was proposed.Finally, the development trend of intelligent wireless communication reception method in the context of signal big data was discussed.It is expected to provide reference for the research and development of wireless communication technology based on deep learning.

    Quantum Internet: key technologies and future prospect
    Zishan LIU, Junsen LAI, Wenyu ZHAO
    2022, 38(2):  18-24.  doi:10.11959/j.issn.1000-0801.2022034
    Asbtract ( 409 )   HTML ( 58)   PDF (509KB) ( 450 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The quantum Internet integrates quantum computing, quantum measurement and communication, which can be described as the future goal of the evolution of quantum information technology.However, due to the restrictions of the laws of quantum mechanics, such as quantum teleportation, quantum entanglement, quantum measurement and non-cloning, which all pose new challenges to the quantum networking design.The basic concepts and development paths of the quantum Internet were firstly introduced, and the implementation differences between the classical network and the quantum network were explained.Considering the difference between quantum communication and the traditional communication, the key technologies for realizing the quantum Internet including the quantum physical devices, networking protocol, quantum decoherence and quantum relay were introduced, and finally the prospect and suggestions for developing quantum Internet was presented.

    Research and Development
    Receiver design of sparse Bayesian learning based MIMO-OFDM power line communication system
    Xinrong LYU, Youming LI, Yongqing WU, Xiaobo TANG
    2022, 38(2):  25-34.  doi:10.11959/j.issn.1000-0801.2022036
    Asbtract ( 145 )   HTML ( 15)   PDF (668KB) ( 89 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The rich impulsive noise in the power line channel poses a huge challenge to the design of MIMO-OFDM transceiver.To solve this problem, a design scheme that can jointly estimate the channel and impulsive noise was proposed, which exploited the parametric sparsity of the classical multipath model and the sparsity of the time domain impulsive noise.In this scheme, the unknown channel model parameters and the impulsive noise were jointly regarded as a sparse vector.By observing the spatial correlation of MIMO system, a compressed sensing model based on multiple measurement vectors was constructed.The multiple response sparse Bayesian learning theory was introduced to jointly estimate the MIMO channel parameters and impulsive noise.The simulation results show that, compared with the traditional receiver scheme that considers MIMO channel estimation and impulsive noise suppression separately, the receiver proposed has a significant improvement in channel estimation performance and bit error rate performance.

    Task offloading and resource allocation in NOMA-based ultra-dense MEC networks
    Yun SHENG, Chen XU, Guangyuan ZHENG
    2022, 38(2):  35-46.  doi:10.11959/j.issn.1000-0801.2022033
    Asbtract ( 251 )   HTML ( 55)   PDF (697KB) ( 265 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In order to improve spectrum efficiency and meet the service demands of massive users, a non-orthogonal multiple access (NOMA)-based ultra-dense mobile edge computing (MEC) system was considered.In order to solve the serious communication interference caused by simultaneous offloading of multiple users and make efficient use of edge server resources, a joint task offloading and resource allocation scheme was proposed to minimize the system energy consumption while meeting the quality of service (QoS) of all users.Offloading decision, power control, computation resource and subchannel resource allocation were jointly considered in the proposed scheme.Simulations results show that the proposed scheme can efficiently lower system energy consumption compared to the other offloading schemes.

    Design and FPGA implementation of an efficient parallel Turbo decoder for combining state metric calculations
    Qian ZHANG, Ming ZHAN, Jianwu ZHANG, Fulong WANG, Yunkai1 FENG, Hao TANG
    2022, 38(2):  47-58.  doi:10.11959/j.issn.1000-0801.2022023
    Asbtract ( 101 )   HTML ( 11)   PDF (838KB) ( 62 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In order to achieve the requirement of high throughput and low-power in wireless communication, a parallel Turbo decoder has attracted extensive attention.By analyzing the calculating of the state metrics, a low-resource parallel Turbo decoder architecture scheme based on merging the forward and backward state metrics calculation modules was proposed, and effectiveness of the new architecture was demonstrated through field-programmable gate array (FPGA) hardware realization.The results show that, compared with the existing parallel Turbo decoder architectures, the proposed design architecture reduces the logic resource of state metrics calculation module about 50%, while the dynamic power dissipation of the decoder architecture is decreased by 5.26% at the frequency of 125 MHz.Meanwhile the decoding algorithm is close to the decoding performance of the parallel algorithm.

    Multi-carrier information hiding algorithm based on three-dimensional model’s concave-convex structure characteristics
    Shuai REN, Lei SHI, Binbin WANG, Huirong CHENG, Qianqian ZHANG, Honglin LIU
    2022, 38(2):  59-70.  doi:10.11959/j.issn.1000-0801.2022008
    Asbtract ( 178 )   HTML ( 19)   PDF (1289KB) ( 104 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the problem that the embedding capacity, invaibility and robustness of a single-carrier information hiding algorithm cannot be further improved due to the limitation of the number of carriers, the voxelization of the carrier and the embedding of secret information were combined with the concave-convex structure characteristics of the three-dimensional model, and a method was proposed.A multi-carrier information hiding algorithm based on the three-dimensional model’s concave-convex structure features.Firstly, the three-dimensional model was voxelized, and the three-dimensional model’s concave-convex structure features were extracted from the data set obtained after voxelization to classify the carrier library, and the concave-convex degree was obtained by conversion after the interval was encoded.Secondly, the secret information was segmented according to the number of carrier classifications and scrambled and optimized, so that the embedding of the carrier and the secret information was effectively connected through its classification and number of segments, and double embedding of secret information by encoding data of concavity intervals and voxelized coordinate points, respectively, to further improve the performance of the algorithm.Finally, the genetic algorithm was applied to optimize the secret information to complete the information hiding.The experiment shows that compared with the high-capacity three-dimensional model steganography algorithm based on a single carrier, the invisibility, robustness and capacity of the algorithm were significantly improved.

    Personalized recommendation model with multi-level latent features
    Qing SHEN, Wenbin GUO, Jungang LOU, Qiangguo YU
    2022, 38(2):  71-83.  doi:10.11959/j.issn.1000-0801.2022035
    Asbtract ( 166 )   HTML ( 23)   PDF (1939KB) ( 430 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Personalized recommendation has become one of the most effective means to solve information overload, and it is also a hot technology in the research field of massive data mining.However, traditional recommendation algorithms often only use the user’s rating information on the item, and lack a comprehensive consideration of the potential characteristics of the user and the item.The factorization machine, wide neural network, crossover network and deep neural network were combined to extract the shallow latent features, low-order nonlinear latent features, linear cross latent features, and high-order nonlinear latent features of users and items.Thus, a new deep learning personalized recommendation model with multilevel latent features was established.The experimental results on four commonly used data sets show that considering the multi-level potential features of users and items can effectively improve the prediction accuracy of personalized recommendations.Finally, the influence of factors such as the dimensions of the embedding layer and the number of neurons on the prediction performance of the new model was studied.

    Time-frequency image and high-order spectrum characteristics based radar signal recognition
    Shitong LI, Daying QUAN, Zeyu TANG, Yun CHEN, Xiaofeng WANG, Xiaoping JIN
    2022, 38(2):  84-91.  doi:10.11959/j.issn.1000-0801.2022024
    Asbtract ( 177 )   HTML ( 14)   PDF (2149KB) ( 107 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Aiming at improving the accuracy of radar signal recognition under a low signal-to-noise ratio, a radar signal recognition algorithm based both on time-frequency image and high-order spectrum feature was proposed.Firstly, the time-frequency image was obtained by Choi-Williams distribution (CWD) transform, based on which the time-frequency image was preprocessed and the texture features were extracted by gray level co-occurrence matrix (GLCM) in sequence.Meanwhile, the symmetrical holder coefficient was used to extract the high-order spectral features of the signal.Then, the texture features and high-order spectrum features were form a new set of joint feature vectors.Finally, with the proposed feature vector the classification and recognition of radar signals were implemented by a support vector machine.The algorithm was verified on the data set with eight typical radar signals.Experimental results show that the recognition accuracy of different radar signals can achieve higher than 90% when the signal-to-noise ratio is -8 dB.

    Video temporal perception characteristics based just noticeable difference model
    Yafen1 XING, Haibing YIN, Hongkui WANG, Qionghua LUO
    2022, 38(2):  92-102.  doi:10.11959/j.issn.1000-0801.2022030
    Asbtract ( 271 )   HTML ( 28)   PDF (2513KB) ( 318 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The existing temporal domain JND(just noticeable distortion) models are not sufficient to depict the interaction between temporal parameters and HVS characteristics, leading to insufficient accuracy of the spatial-temporal JND model.To solve this problem, feature parameters that can accurately describe the temporal characteristics of the video were explored and extracted, as well as a homogenization method for fusing heterogeneous feature parameters, and the temporal domain JND model based on this was improved.The feature parameters were investigated including foreground and background motion, temporal duration along the motion trajectory, residual fluctuation intensity along motion trajectory and adjacent inter-frame prediction residual, etc., which were used to characterize the temporal characteristics.Probability density functions for these feature parameters in the perception sense according to the HVS(human visual system) characteristics were proposed, and uniformly mapping the heterogeneous feature parameters to the scales of self-information and information entropy to achieve a homogeneous fusion measurement.The coupling method of visual attention and masking was explored from the perspective of energy distribution, and the temporal-domain JND weight model was constructed accordingly.On the basis of the spatial JND threshold, the temporal domain weights was integrated to develop a more accurate spatial-temporal JND model.In order to evaluate the performance of the spatiotemporal JND model, a subjective quality evaluation experiment was conducted.Experimental results justify the effectiveness of the proposed model.

    Customer service complaint work order classification based on matrix factorization and attention multi-task learning
    Yong SONG, Zhiwei YAN, Yukun QIN, Dongming ZHAO, Xiaozhou YE, Yuanyuan CHAI, Ye OUYANG
    2022, 38(2):  103-110.  doi:10.11959/j.issn.1000-0801.2022031
    Asbtract ( 227 )   HTML ( 39)   PDF (464KB) ( 207 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The automatic classification of complaint work orders is the requirement of the digital and intelligent development of customer service of communication operators.The categories of customer service complaint work orders have multiple levels, each level has multiple labels, and the levels are related, which belongs to a typical hierarchical multi-label text classification (HMTC) problem.Most of the existing solutions are based on classifiers to process all classification labels at the same time, or use multiple classifiers for each level, ignoring the dependence between hierarchies.A matrix factorization and attention-based multi-task learning approach (MF-AMLA) to deal with hierarchical multi-label text classification tasks was proposed.Under the classification data of real complaint work orders in the customer service scenario of communication operators, the maximum Top1 F1 value of MF-AMLA is increased by 21.1% and 5.7% respectively compared with the commonly used machine learning algorithm and deep learning algorithm in this scenario.It has been launched in the customer service system of one mobile operator, the accuracy of model output is more than 97%, and the processing efficiency of customer service agent unit time has been improved by 22.1%.

    Research on simulation of 5G indoor distribution system based on ray tracing propagation model
    Xin TAO, Junqing WAN, Yifeng LI, Xisheng XU, Nanting QIAO, Bin XU
    2022, 38(2):  111-118.  doi:10.11959/j.issn.1000-0801.2022027
    Asbtract ( 328 )   HTML ( 37)   PDF (25976KB) ( 145 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The construction of indoor distribution system has become an effective means to solve the problem of deep coverage.At present, 5G network construction is developing rapidly, indoor distribution system construction is exploding, and the network planning and design of indoor distribution system needs to be more refined.How to effectively evaluate the indoor signal coverage effect becomes one of the important tasks of current indoor distribution system planning and design.A building was taken as the research object, the coverage performance index was calculated by ray-tracing propagation model simulation, and the results were compared with the road test data of the existing network.After correcting the relevant model parameters, the simulation is consistent with the actual measurement, and the demonstration simulation provides a reference basis for network deployment.

    Research on adaptive optimization of KNN-DBSCAN parameters based on MDT overlapping coverage data
    Lu LIU, Ruijie CHEN, Jia LI
    2022, 38(2):  119-129.  doi:10.11959/j.issn.1000-0801.2022010
    Asbtract ( 218 )   HTML ( 18)   PDF (1880KB) ( 264 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In the traditional network optimization, the drive test (DT) work has obvious disadvantages, such as difficult to fully test roads and buildings, large test workload, low work efficiency, long cycle, affected by human factors, unable to dynamically pay attention to the network quality of each area, and the conventional measurement report (MR) data does not have positioning information, so it is impossible to accurately locate the location where the overlapping coverage problem occured.Based on minimization drive test (MDT), the precision positioning system collected the MDT data of the underlying base station and outputted the grid with high overlapping coverage according to the overlapping coverage algorithm.Then, the sensitivity of DBSCAN algorithm to parameter setting was solved through the adaptive K-nearest neighbor density-based spatial clustering of applications with noise (KNN-DBSCAN)joint algorithm.The problem grid was unsupervised clustered, the problem contiguous area was converged, and the grid area was mapped through the cell sampling contribution.Finally, the global top cell was accurately adjusted to optimize the high overlap coverage.

    Engineering and Application
    Research on access schemes of SD-WAN multi-cloud aggregation platform
    Jia NIU, Yongming YAN, Qiangen ZHAO, Jiajun MIN, Songyang QI
    2022, 38(2):  130-138.  doi:10.11959/j.issn.1000-0801.2022020
    Asbtract ( 325 )   HTML ( 32)   PDF (2106KB) ( 188 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    A detailed description of the current situation, application, demand analysis, basic functions and application scenarios of the multi-cloud aggregation platform were given, and the advantages of SD-WAN compared with the traditional private line multi-cloud access mode was discussed.The networking projects of multiple SD-WAN accessing multi-cloud aggregation platforms were analyzed, a multi-cloud access scheme combining SD-WAN and dedicated line to realize network link monitoring were proposed.The network quality monitoring function of SD-WAN access to multi-cloud and the monitoring function of SD-WAN on underlay dedicated line under the fusion networking was proposed.Based on the topology design of network architecture combining traditional private line to multi-cloud and segmentalized access mode of SD-WAN and traditional private line to multi-cloud, the application scenarios of hybrid network to multi-cloud was introduced and the value of SD-WAN to multi-cloud was reflected from the perspective of fusion network by combining advantages and disadvantages.

    A solution for 5G indoor and outdoor co-frequency network interference
    Xirui GUO, Tao ZHANG, Qiang ZHANG, Dongyang WANG, Yan YANG
    2022, 38(2):  139-148.  doi:10.11959/j.issn.1000-0801.2022029
    Asbtract ( 514 )   HTML ( 33)   PDF (1425KB) ( 257 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    5G indoor and outdoor deployment with the same frequency will lead to the decline of network coverage performance and user service experience.The causes of indoor and outdoor interference with the same frequency was systematically analyzed.The principle of three indoor and outdoor co-channel interference solutions, namely PRB random interference solution, indoor and outdoor beam cooperative interference solution and indoor multi beam interference solution were studied.The results show that the PRB random interference solution has the best interference suppression effect when the traffic load is about 30%.When the indoor and outdoor beam cooperation function is turned on when the adjacent area of the outdoor macro station is no-load or high load (70%), the improvement of the downlink rate is more obvious.When the indoor multi-beam function is turned on, the downlink rate is significantly higher than that of single beam.Finally, the deployment suggestions of co-channel interference control scheme were given.

    Research and application of UMR technology for new metropolitan area network
    Yuanyang YIN, Zhizhong WANG, Youqiang JIANG, Jiaqi SUN, Xiaobin LIANG
    2022, 38(2):  149-156.  doi:10.11959/j.issn.1000-0801.2022028
    Asbtract ( 370 )   HTML ( 53)   PDF (864KB) ( 241 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The new metropolitan area network adopts the new architecture of spine-leaf.BRAS is a high hanging and centralized deployment mode.With the deployment of the new metropolitan area network, the home wide service is gradually cut over to the new network.The current typical bearer scheme has many problems, such as the sharp increase of pressure on the network equipment MAC routing table, the large and complex network configuration.After analyzing and comparing the defects of typical bearer schemes, the implementation scheme of UMR technology was innovatively put forward, which provided reference significance for promoting the rapid deployment of new metropolitan area network.

    AI engineering technology solutions in network intelligence
    Mingwei ZHU
    2022, 38(2):  157-165.  doi:10.11959/j.issn.1000-0801.2022016
    Asbtract ( 345 )   HTML ( 61)   PDF (557KB) ( 471 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Depending on AI technology, network intelligence is becoming an important initiative for communication industry to enhance network empowerment externally, and to achieve cost reduction and efficiency internally.The difficulties implementing network intelligence applications from the perspective of AI engineering were analyzed.The industrial grade AI engineering technical solutions were proposed, including data collection and processing, computing resources management and task scheduling, and inference deployment optimization.The strategies of network intelligence’s ecosystem development were studied.

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: