Current Issue

• • 2019, Vol. 42 No. 3 Published date:28 June 2019 Last issue Next issue

  Papers

  • A SCMA Codebook Design Based on Distance Spectrum
  • GUO Ming-kun, SHAO Xiao-tao, YANG Wei
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 1-6,28. DOI:10.13190/j.jbupt.2018-153
  • Abstract ( 431 )     HTML     　　
  • Aiming at the problem that the complexity of multidimensional constellation design, and the difficulty of determining optimal constellation operators actually increase the difficulty of codebook design in the typical sparse code multiple access(SCMA) codebook design scheme, which gets user codebooks from a multidimensional constellation via the constellation operation, the influence of user codebooks on the specific process of message passing algorithm(MPA) is analyzed, and a codebook design scheme based on maximizing the minimum element of the distance spectrum is proposed. The scheme transforms the process of multidimensional constellation design, and constellation operators determination into a simple codebook design parameter selection process. Without affecting the unique decodability of user codebooks, the complexity of parameter selection is reduced by replacing some codebook design parameters. Simulations show that compared with the existing SCMA codebook design schemes, the scheme based on distance spectrum can significantly improve the bit error rate performance of system.
  • References | Supplementary Material | Related Articles

  • Low-Complexity MMSE Demodulation Algorithm for GFDM
  • HUANG Xiang-dong, WANG Hui-jie, LI Ming-shi, CAO Yue-bin
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 7-13. DOI:10.13190/j.jbupt.2018-157
  • Abstract ( 386 )     HTML     　　
  • Aiming at the problem that the computation complexity of the existing minimum mean square error (MMSE) demodulator with generalized frequency division multiplexing (GFDM) over frequency selective channel is relatively high, a matrix deconstruction based low-complexity MMSE GFDM receiver is proposed. By utilization of the particular structures (sparse, quasi-tridiagonal, block-symmetric, etc.) in a series of large-sized matrices involved in the receiver, either of large-sized matrix manipulations (including matrix multiplication and matrix inversion) is partitioned into small-sized matrix manipulations, thereby considerably reducing the computational complexity. Quantitively speaking, the times of complex multiplications consumed by the proposed MMSE receiver is 2-3 amount levels lower than the original MMSE receiver. Moreover, simulation shows that the proposed receiver does not suffer from any bit-error-rate degradation. Therefore, the proposed MMSE GFDM receiver possesses vast potentials for future 5G demodulation applications.
  • References | Supplementary Material | Related Articles

  • Queue-Aware Joint Optimization of Offloading and Transmission in Wireless Mobile Edge Computing Systems
  • TENG Ying-lei, LIU Wei, OUYANG Wei-ping, LI Kun, SONG Mei
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 14-20. DOI:10.13190/j.jbupt.2018-256
  • Abstract ( 511 )     HTML     　　
  • Considering task queue model on both user and mobile edge computing (MEC) server side, a dynamic computing task offloading problem in multi-user-multi-MEC-server system is proposed. To find the optimal offloading and corresponding uplink precoding strategy, a long-term average overhead containing delay and power consumption of the whole system optimization problem is formulated. The original problem is transformed into a single-stage cost target optimization problem based on Lyapunov optimization method and further converted into a typical DC programming problem by utilizing successive approximation and semi-definite relaxation. Simulation shows that the proposed scheme characterized by optimizing precoding design can meet the lower delay and power consumption requirements.
  • References | Supplementary Material | Related Articles

  • English Textual Entailment Recognition Using Capsules
  • ZHU Hao, TAN Yong-mei
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 21-28. DOI:10.13190/j.jbupt.2018-289
  • Abstract ( 442 )     HTML     　　
  • An English textual entailment recognition method using capsules is presented. This method builds a capsule for each relationship to model the recognition of this relationship and assigns it as the attribute of the capsule. Given two texts, they are first encoded by highway encoding layer and sequence encoding layer to obtain the semantic representations, and then fed into each capsule, passing through its internal interaction module, comparison module and aggregation module in turn. The interaction module uses the inter-attention mechanism to extract local interactive features between texts. The comparison module and the aggregation module use the feedforward network to compare and aggregate the semantic information. Finally, the output of all capsules is normalized to obtain the relationship between the two texts. The accuracy on SNLI test dataset is 89.2%, the accuracy on MultiNLI matched and mismatched test dataset is 77.4% and 76.4%. The visual analysis of the attentional relationship matrix of interaction module also verifies the effectiveness of capsules in the English textual entailment recognition task.
  • References | Supplementary Material | Related Articles

  • A Secure Clustering Routing Protocol Based on Trust and Balancing Energy Consumption
  • PAN Lei-na, TAO Yang, XU Xiang-yang, WANG Jin
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 29-36. DOI:10.13190/j.jbupt.2018-158
  • Abstract ( 432 )     HTML     　　
  • Aiming at the problems that the trust model is simple and the energy consumption is not optimized enough in the existing trust-aware routing protocols. A secure clustering routing protocol based on trust and balancing energy consumption (SCR-TBE) is proposed. Introducing multiple evaluation factors, direct trust is calculated by fuzzy comprehensive evaluation model, and the recommendation trust is filtered and weighted allocation based on the deviation degree, and comprehensive trust is calculated added historical trust value. And the cluster head penalty coefficient is designed to speed up the identification and isolation of malicious cluster heads. By dividing the monitoring area, the unequal clustering is achieved to alleviate the energy hole phenomenon. The node trust value, dump energy and data transmission distance are chosen as routing factors. The cluster function and forwarding function are designed to reduce the chance of malicious nodes participating in data transmission. Simulation results show that, compared with LEACH-C and TLES protocols, both the number of data packets and the energy consumption balance have been greatly improved, enhancing the security and reliability of wireless sensor networks.
  • References | Supplementary Material | Related Articles

  • Task Transition Probability Based Anomaly Detection Method for Sensor Nodes
  • MA Jun-yan, ZHANG Te, WANG Jin
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 37-42. DOI:10.13190/j.jbupt.2018-114
  • Abstract ( 467 )     HTML     　　
  • The anomaly detection of sensor nodes is a great challenge to wireless sensor networks. A feature based on task transition probability is therefore proposed to model running states of sensor nodes, and the feature can be further used for anomaly detection. Task transition probability based anomaly detection (T2PAD) analyzes states of sensor nodes based on the one-step transition probability of running tasks within the nodes, and then performs anomaly detection by comparing similarities between transition probability vectors. T2PAD can identify tasks that caused the anomaly and narrow down the scope of problematic code, which provides clues to deal with the anomaly. Case studies on defects from a sensor network open source project are carried out to verify the effectiveness of T2PAD.
  • References | Supplementary Material | Related Articles

  • In-Depth Recognition of Human Motion States Based on Smart Phone Perception
  • YIN Xiao-ling, XIA Qi-shou, CHEN Xiao-jiang, HE Juan, CHEN Feng
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 43-50. DOI:10.13190/j.jbupt.2018-221
  • Abstract ( 749 )     HTML     　　
  • In order to improve the accuracy of recognition of human motion states by smart phones, an in-depth recognition method based on parallel convolution neural network (PCNN) is proposed. Firstly, the sensor data input format is standardized by using 3D data matrix. Secondly, two PCNNs are used to carry out convolution and pool operation to the acceleration sensor and gyroscope data of human body motion respectively, realizing partial weight sharing. Finally, the two PCNNs are merged in the full-connected layer, and the softmax function is used to classify the human motion states. Experiments show that this method can extract the deep features of human motion states from the original data of the sensor, which can improve the recognition rate of the motion state by comparing with the traditional machine learning method.
  • References | Supplementary Material | Related Articles

  • An Asymmetric Feed Compact High Isolation Diversity Antenna
  • YANG Yang, LIU Yuan-an, WU Fan
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 51-57. DOI:10.13190/j.jbupt.2018-245
  • Abstract ( 400 )     HTML     　　
  • A high isolation compact diversity antenna structure based on asymmetric coplanar band (ACS) feed is proposed. Use of ACS as the antenna feed structure reduces the antenna volume by about 50%; by introducing the floor extension branch as a decoupling structure, the high antenna port isolation is achieved to 15 dB and is highest to 30 dB. It is shown that the diversity antenna can satisfy the 2.4/5.2 GHz WLAN, 2.6/3.5 GHz LTE communication requirements.
  • References | Supplementary Material | Related Articles

  • Recognition and Application of Abnormal Sound Via SVM Based on PSO-PF
  • WEI Juan, ZHANG Peng-nan, YUE Feng-li, NING Fang-li
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 58-63. DOI:10.13190/j.jbupt.2018-246
  • Abstract ( 451 )     HTML     　　
  • In order to solve the problems of low recognition accuracy and high computation complexity in abnormal sound signals, a particle filter based on particle swarm optimization (PSO-PF) algorithm is proposed to optimize the parameters of support vector machine (SVM). To improve the estimation precision of particle filter, the particle swarm optimization is applied to drive all the particles to the regions in which their likelihoods are high, by updating the velocity and position of particles constantly. And the algorithm can avoid falling into local optimum in SVM parameter optimization. The experimental results show that the new algorithm can achieve higher recognition accuracy and lower computation complexity for abnormal sounds recognition.
  • References | Supplementary Material | Related Articles

  • Research on Adaptive Strategy of End Hopping System Based on Deep Belief Nets
  • SHI Le-yi, LI Jian-lan, GUO Hong-bin, MA Meng-fei, CHEN Hong-long
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 64-71. DOI:10.13190/j.jbupt.2018-268
  • Abstract ( 452 )     HTML     　　
  • For problem of single hopping strategy in active cyber defense of end hopping, the multiple and heterogeneous hopping patterns are introduced into the end hopping system, the definition of end information is expanded, and the self-adaptive adjustment scheme of the end hopping is given. Further, an adaptive model of end hopping based on deep belief network is proposed, and the process of data collection, feature extraction and state prediction are formalized. The state feature index of end hopping network is defined and the characteristic data set is established. The Markov chain is used to predict the state of network in the next cycle. The heterogeneous hopping mode is selected according to the prediction results. Thus, the adaptive change of the end hopping mode is realized. Experiments show that the network state recognition and prediction of the model all have high accuracy, and the heterogeneous hopping strategy can effectively resist different types of attacks, which illustrates the validity and security of the end hopping adaptive model.
  • References | Supplementary Material | Related Articles

  • JR-CoMP-Based Clustering and Resource Allocation Scheme in UDN
  • TIAN Xin-ji, ZHANG Hao-tian, LI Xiao-jing, ZHANG Chang-sen
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 72-78. DOI:10.13190/j.jbupt.2018-208
  • Abstract ( 445 )     HTML     　　
  • For uplink interference problem in ultra dense networks(UDN), a joint clustering and resource allocation scheme is proposed based on coordinated multiple point joint reception(JR-CoMP). Firstly, users and small base stations are divided into several clusters according to the uni-subband clustering. The multiple edge users in the same small cell are assigned to different user clusters. The users at the close edge of the adjacent small cell are assigned to the same cluster. The cell-center users are clustered according to the interference intensity of them to the existing base station clusters. Then, multi-sub band allocation is performed for user clusters. The user clusters containing the same small cell users are allocated different sub-bands, and the number of allocated sub-bands for each user clusters is as much as possible. Finally, the water-filling algorithm is used to allocate appropriate power to each user with the goal of maximizing the total transmission rate of the system. Simulation results show that spectral efficiency can be significantly improved by the proposed scheme.
  • References | Supplementary Material | Related Articles

  • A Modulation Format Identification Method for Probabilistic Shaping Optical Transmission System
  • LI Zhi-pei, WANG Xi-shuo, LIU Bo, ZHANG Qi, XIN Xiang-jun
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 79-82,126. DOI:10.13190/j.jbupt.2018-014
  • Abstract ( 405 )     HTML     　　
  • A blind modulation format identification method based on fourth-power eigenvalues is proposed for variable-rate optical transmission systems employing probabilistic shaping technology and high-order modulation formats. By using the fast Fourier transformation (FFT) peak value and variance after the received signal passes through fourth-power operation, accurate identification of high order modulation formats with probabilistic shaping and polarization multiplexing can be realized. A high-speed long-haul coherent optical transmission system is built on the platform of the joint simulation to verify the effectiveness of the method.
  • References | Supplementary Material | Related Articles

  Reports

  • An Information Diffusion Pattern Mining Method Based on Communication Actions
  • XIANG Ying-zhuo, WEI Qiang, YOU Ling
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 83-90. DOI:10.13190/j.jbupt.2018-204
  • Abstract ( 439 )     HTML     　　
  • To deal with the challenges of information diffusion pattern mining problem which the communication content is unknown and innocent data occupies a very high ratio of the observed data, the article proposes a probability model predicting the relativity of the communications between users, which infers the information diffusion. In addition, it proves the inferring problem NP-hard, and proposes NetMine algorithm to get a near optimal solution. Experiments show that the proposed NetMine algorithm outperforms other state-of-art algorithms.
  • References | Supplementary Material | Related Articles

  • Modeling the Statistical Distribution of the Inter-Contact Times for Communication Opportunities in Vehicle Networks
  • HUANGFU Wei, YANG Xin-zhu, WANG Huan, HU Xiao-yan
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 91-97. DOI:10.13190/j.jbupt.2018-230
  • Abstract ( 447 )     HTML     　　
  • As a kind of opportunistic networks, the vehicle network, in which vehicles communicate instantaneously with each other only when they meet opportunely, attracts more and more attentions. The inter-contact times of communication opportunities depend on the factors such as the vehicle moving model and the vehicle density. Based on the data of the taxi traces in Beijing city, an efficient method which combines the GeoHash-based filter and the trace interpolation is proposed to extract all the instants of the communication opportunities to model the statistical distribution. It is concluded that the inter-contact times obey an exponential distribution if the interval is small while a log-normal one if the interval is large. Such a piecewise heavy-tailed distribution will affect the design of the vehicle networks to some extent.
  • References | Supplementary Material | Related Articles

  • An Iteration Localization Algorithm in Wireless Sensor Network Based on Intersection Degree Ratio
  • QIAN Kai-guo, PU Chun-fen, WANG Yu-jian, SHEN Shi-kai
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 98-105. DOI:10.13190/j.jbupt.2018-233
  • Abstract ( 441 )     HTML     　　
  • In order to improve the performance of range-free localization algorithm such as DV-Hop and Amorphous for wireless sensor network,the problem of using the minimum hop path distance to replace the Euclidean distance between nodes and choosing the approximate collinearity beacons to calculate location of unknown node that introduces large positioning error is studied. An iteration localization algorithm in wireless sensor network based on intersection degree ratio (ILA-IDR) is proposed. The ILA-IDR algorithm first uses the localization group topology distribution quality function to select three 1-hop neighbor reference nodes form a non-collinear positioning unit, and then the distance calculation based on the intersection degree ratio is used to improve the estimation distance accuracy. Finally, the hyperbolic calculation method is adopted to reduce the error to improve the node location accuracy. Simulation shows that the ILA-IDR algorithm is superior to those of the DV-Hop, Amorphous and the advanced algorithm for that the localization error of the ILA-IDR algorithm is lower than those algorithms both in the node uniform and C-Shape distribution scenario.
  • References | Supplementary Material | Related Articles

  • High Performance Row-Based Hashing GPU SpGEMM
  • TANG Yang, ZHAO Da-fei, HUANG Zhi-bin, DAI Zhi-tao
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 106-113. DOI:10.13190/j.jbupt.2018-252
  • Abstract ( 820 )     HTML     　　
  • Aiming at the performance problem of general sparse matrix-matrix multiplication (SpGEMM), a graphics processing unit (GPU)-accelerate SpGEMM algorithm based on task classification and low-latency Hashing table, RBSPARSE, was presented in the paper. RBSPARSE consists of a low-cost pre-analysis method to identify the complexity of sub-tasks, and a Hashing table-based algorithm which could utilize low-latency shared memory to achieve max efficiency. By taking the load balancing issue and the memory latency issue into consideration, RBSPARSE could significantly reduce the overall time in computation. RBSparse and BHSparse are compared. BHSparse is the previous state-of-the-art algorithm for SpGEMM. The result shows that our algorithm is 3.1 times faster than BHSparse on average, and could achieve a maximum 14.49 times faster speed in the best scenario.
  • References | Supplementary Material | Related Articles

  • Word Sense Disambiguation Based on Convolution Neural Network
  • ZHANG Chun-xiang, ZHAO Ling-yun, GAO Xue-yao
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 114-119. DOI:10.13190/j.jbupt.2018-148
  • Abstract ( 593 )     HTML     　　
  • In order to improve the performance of word sense disambiguation (WSD), a disambiguation method based on convolution neural network (CNN) is proposed. Ambiguous word is viewed as center and four adjacent word units around its left and right sides are extended. Word, part-of-speech and semantic categories are extracted as disambiguation features. Based on disambiguation features, CNN is used to determine semantic categories of ambiguous words. Training corpus of SemEval-2007:Task#5 and semantic annotation corpus from Harbin Institute of Technology are used to optimize CNN classifier. Testing corpus of SemEval-2007:Task#5 is used to test the performance of WSD classifier. Average disambiguation accuracy of the proposed method is improved. Experiments show that this method is feasible in WSD.
  • References | Supplementary Material | Related Articles

  • A Beam Selection Method Based on Unmanned Aerial Vehicle Millimeter Wave Communication
  • MU Xi-dong, GUO Li, DONG Chao, LIN Jia-ru
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 120-126. DOI:10.13190/j.jbupt.2018-277
  • Abstract ( 500 )     HTML     　　
  • A millimeter-wave unmanned aerial vehicle (UAV) cellular multiuser network is investigated based on beamspace multiple-input and multiple-out put. A beam selection scheme assisted by UAV mobility is proposed. At first, an optimal deployment position of UAV base station is identified with the aim of reducing power leakage problem among users. Afterwards, a simple power focused beam selection criterion called beam reselection is introduced to avoid the problem that the same beam is selected by different users, which will cause the waste of radio-frequency chains. Numerical simulations results show that the proposed scheme achieve near optimal achievable sum rate performance and higher power efficiency compared with conventional beam selection scheme.
  • References | Supplementary Material | Related Articles

  • A Fast Feature Selection Framework and Method
  • QIU Li-ke, LIU Jing, SUN Zhong-wei, ZHAO Yang-fan
  • Journal of Beijing University of Posts and Telecommunications. 2019, 42(3): 127-132. DOI:10.13190/j.jbupt.2018-151
  • Abstract ( 542 )     HTML     　　
  • Aiming at the imbalance between accuracy and computational efficiency in feature selection, a fast feature selection framework (FFFS) is proposed. Based on this framework, a fast feature selection algorithm, MRMR-SFS, is proposed. The minimum redundancy maximum relevance (MRMR) method is used to select the candidate features, and sequential forward selection (SFS) method is used to verify the performance of the candidate features as well. It improves the calculation efficiency by limiting the number of iterations. Comparison experiments with the MRMR, SFS and a filter-dominating hybrid sequential floating forward selection algorithms demonstrate that MRMR-SFS can balance the accuracy and computational efficiency well.
  • References | Supplementary Material | Related Articles