Please wait a minute...

Journal of Beijing University of Posts and Telecommunications

  • EI核心期刊

Current Issue

    • Human Trajectory Prediction by Multi-resolution Interaction
    • LIU Shaohua, SUN Jingkai, WANG Yisu, LIU Haibo, MAO Tianlu
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 1-6.
    • Abstract ( 388 )       
    • Human trajectory prediction is an essential and challenging task in robot navigation and autonomous driving applications. One of the most challenging tasks is to model the interaction between pedestrians. Pairwise attention is used by most of the existing models to model the interaction. However, when there are too many pedestrians in the scene, these methods have redundancy in interaction modeling and ignore the interaction differences of pedestrians at different distances. To address these challenges, a multi-resolution global-local model is proposed, which contains a novel multi-region interaction sub-network to capture the global interaction and an additional local interaction sub-network to model pedestrians' interactions in the local neighborhood. In the meantime, the temporal attention mechanism is introduced in the proposed model to fuse the interactive information of different time steps. The experimental results show that compared with previous models, the proposed model achieves better performance on two publicly available datasets.
    • Supplementary Material | Related Articles
    • Research on Airport Clearance Model and Its Analytical Formula of Variable-Slope Runway
    • ZHUANG Yufeng, ZHANG Ningxi, CHEN Xiying, HAN Jingwen
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 7-15.
    • Abstract ( 327 )       
    • A model of airport clearance with a variable-slope runway is proposed, and its analytical formula is studied to improve the efficiency and accuracy of building super elevation evaluation in the airport clearance area. First, an airport clearance 3D model is established according to the airport runway parameters. The plane limitation range of each limit surface is calculated by considering the influence of runway slope and airport grade on the obstacle limit surface. Then, the clearance elevation model is constructed, and an algorithm is designed to determine the type of limit surface where the point is on, and calculate the limit height. The constructed airport clearance 3D model and elevation model are applied to the actual study, which verifies the validity of the model, and the accuracy and speed of the clearance limit height calculation are improved. Furthermore, the visualization of the clearance model is realized based on Cesium, which is convenient for the staff to quickly judge whether the building exceeds the clearance limit.
    • Supplementary Material | Related Articles
    • Performance Analysis of Vehicular Edge Computing Based on Stochastic Network Calculus
    • LI Song, WANG Xinrong, WANG Bowen, SUN Yanjing, CHEN Ruirui
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 16-22.
    • Abstract ( 338 )       
    • To analyze the performance of task offloading in Internet of vehicles, the task delay and data backlog of edge computing of Internet of vehicles millimeter-wave communication is analyzed based on moment generating function method of stochastic network calculus. First, the mathematical models of the arrival process of vehicular tasks, the service process of millimeter-wave communication and the edge computing is established, and the moment generating functions of these process are derived. Then, the closed-form solution of delay and backlog probability bounds of the vehicular computing task offloading is obtained. Finally, the theoretical analysis is verified by the Monte Carlo simulation.
    • Supplementary Material | Related Articles
    • Joint Beamforming Scheme Based on PSO Algorithm for RIS-MIMO System
    • ZHENG Feng, YANG Li
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 23-29.
    • Abstract ( 783 )       
    • To maximize the spectral efficiency of the multiple-input multiple-output ( MIMO) system assisted by reconfigurable intelligent surface (RIS), the active beamforming of base station and the phase shift design of RIS have to be considered jointly. Therefore, a beamforming scheme based on the improved particle swarm optimization (PSO) algorithm is proposed. First, the joint beamforming problem is decomposed into active beamforming at the base station and phase shift design at RIS by using an alternating optimization algorithm. Then, the two problems are solved alternately by the water-filling algorithm and improved PSO algorithm. Finally, the proposed scheme is extended to the RIS of the practical phase-shift model. The experimental results show that the beamforming problem can be effectively solved by the proposed scheme, and the spectral efficiency of the system is greatly improved.
    • Supplementary Material | Related Articles
    • An Evaluation of Power Battery Discharge Performance Based on Electric Vehicle Driving Data
    • LYU Zheqing, XIU Jiapeng, YANG Zhengqiu, LIU Yanxin
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 30-35.
    • Abstract ( 614 )       
    • With the wide application of power batteries in new energy vehicles, problems such as battery failure diagnosis and retired battery evaluation have to be solved urgently. Evaluating the state-of-health is a significant topic in the power battery field. Based on the analysis of the driving data of electric vehicles,a model is proposed which can evaluate the discharge performance of lithium-ion power batteries. According to the randomness of charge-discharge data in battery work, the adjustment and adaptation are carried out in the proposed model to solve the problem that the traditional evaluation method is difficult to apply to the real scenario. Due to its low complexity, it can give the evaluation results quickly, which is suitable for mass battery evaluation. Experiments show that the discharge performance index extracted by the proposed model decreases as the cumulative mileage increases, which has a similar trend with the change of battery capacity with mileage according to the grey relational analysis. It can evaluate the current discharge capacity of the battery from the actual work circumstance, and provide a reference for evaluating the state-of-health comprehensively.
    • Supplementary Material | Related Articles
    • 3D Ray Reconstruction Method Based on Enhanced CVAE
    • ZHU Jun, YANG Jun, LI Kai, YU Wenxin
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 36-41.
    • Abstract ( 238 )       
    • The incomplete sample space of ray-tracing-data may increase high-prediction-error users in the massive multiple-input multiple-output channel amplitude prediction. To characterize the channel propagation features of all users, a method for 3D ray reconstruction is proposed based on extended probability distribution conditional variational auto-encoder (CVAE). The prior probability distribution is selected based on the sparsity of user ray samples. A new training set of ray samples is generated for high-prediction-error users by enhancing CVAE to make the latent variable distribution of ray-tracing-data fit the features of high-prediction-error users better. The simulation results show that the number of high-prediction-error users can be reduced to 53.59% by new training set based on the proposed method. Moreover, the new set improves the channel amplitude prediction accuracy by 7.8% while significantly reducing the time overhead of predicting the channel amplitude.
    • Supplementary Material | Related Articles
    • Defect Recognition of Printed Circuit Board Based on YOLOv3-Dense
    • YANG Jie, ZHANG Shujie
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 42-48.
    • Abstract ( 333 )       
    • Accurate detection of tiny defects in printed circuit board processing is the prerequisite to ensure the quality of electronic products. Due to the small feature size and complex circuit layouts, the existing target detection methods have many shortcomings. To solve this problem, a YOLOv3-dense target detection model is proposed for printed circuit board defect detection based on YOLOv3 algorithm. First, dense connection network modules are used to replace some residual units in the feature extraction network so as to enhance feature reuse of the network. Then, the loss function is improved, and the generalized intersection ratio between the prediction box and the true value is used to solve the problem that the optimization cannot continue when the intersection ratio is zero. The experimental results show that compared with other models, the proposed model can improve the recognition accuracy and reduce the model size.
    • Supplementary Material | Related Articles
    • Continuous Sign Language Recognition Based on CM-Transformer
    • YE Kang, ZHANG Shujun, GUO Qi, LI Hui, CUI Xuehong
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 49-53,78.
    • Abstract ( 459 )       
    • To capture the global and local features of sign language actions and preserve the original structure and context in the image, an improved convolution multilayer perceptron Transformer ( CM-Transformer) model is proposed for continuous sign language recognition. The structural consistency advantage of convolution layer and the global modeling performance of self attention model encoder are combined by CM-Transformer to capture long-term sequence dependence. Meanwhile, the feedforward layer of self attention model is replaced by multilayer perceptron to perform translation invariance and locality. In addition, random frame discarding and random gradient stopping techniques are used to reduce the training computation in time and space, and prevent over fitting. Thus, an efficient and lightweight network has been constructed. Finally the connectionist temporal classification decoder is used to align the input and output sequences to obtain the final recognition result. Experimental results on two large benchmark data sets show the effectiveness of the proposed method.
    • Supplementary Material | Related Articles
    • Construction and Performance Analysis of SWIPT-Based Product Polar Coded Cooperation
    • ZHANG Shunwai, XIA Zihan
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 54-59.
    • Abstract ( 177 )       
    • In order to achieve green reliable communication, a product polar coded cooperation system based on simultaneous wireless information and power transfer (SWIPT) is proposed. Firstly, the product polar coded cooperation model based on SWIPT is established. Secondly, the polar code is used as the component code of the product code to construct the product polar code, which is row coded at the source node and column coded at the relay node. Finally, a two-step decoding algorithm with low decoding delay is proposed at the destination node. In the first step, row successive cancellation ( SC) decoding and column SC decoding are performed on the codeword matrix of the received product polar code. If the first step decoding fails, the second step decoding is performed, in which the codeword matrix of the product polar code is transformed into a single line polar code, and then decoded by SC. Theoretical analysis and simulation show that compared with point-to-point system, the product polar coded cooperation based on SWIPT greatly reduces the decoding delay. Compared with the product low density parity check coded cooperative system that uses the belief propagation decoding algorithm, the proposed system has better bit error performance.
    • Supplementary Material | Related Articles
    • Dispersion Relations and Heat Capacity with Long Range Interactions in a Square Lattice
    • LI Zhendong, TAN Zhenting, FANG Ping
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 60-65.
    • Abstract ( 270 )       
    • The dispersion relation of a monoatomic two-dimensional lattice with long-range interactions is derived according to the theory of lattice dynamics. Based on a simple harmonic approximation, a general dispersion relation that takes into account arbitrary neighbor interactions is obtained. The results of the nearest neighbor and the second nearest neighbor approximation are analyzed. When the order of the nearest neighbors p is large, the contribution of the pth nearest neighbors to the average vibration frequency in the Brillouin zone decays in a power law form with the increase of the neighbor number. Moreover, the long-range Coulomb interaction is analyzed as an example. Based on this microscopic dynamics, the effect of long-range interaction on macroscopic heat capacity is numerically investigated, and the heat capacity of the system decays in a power law form with the increase of the neighbor number.
    • Supplementary Material | Related Articles
    • Low-Frequency Wireless Communication Mechanical Antenna Based on Piezoelectric Effect
    • YANG Jing, WEN Xi, YANG Shaolong, PAN Xuchao, BI Ke
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 66-71.
    • Abstract ( 563 )       
    • The size of traditional electric antenna is proportional to the transmitted wavelength. Although the low-frequency electromagnetic wave (30 ~ 300 kHz) reaches several kilometers, the low-frequency electric antenna has some problems, such as excessively large size, poor concealment and high power consumption. To solve the problems of large size and power consumption of electrical antenna and improve the flexibility of low-frequency wireless communication applications, a mechanical antenna based on piezoelectric effect is proposed. Mechanical antenna generates electromagnetic waves by direct excitation of mechanical motion charges. The physical mechanism of the piezoelectric mechanical antenna is studied. The impedance characteristics of piezoelectric mechanical antenna are studied by impedance frequency curve. The low-frequency radiation performances of piezoelectric mechanical antennas with various side lengths are compared by using a wireless communication system. Meanwhile, the magnetic field radiated by piezoelectric mechanical antenna changes with the distance between the receiving end of wireless communication system and piezoelectric mechanical antenna, and its variation law is verified. The result shows that the piezoelectric mechanical antenna has great potential in portable, low-cost and high-performance low-frequency wireless communication applications.
    • Supplementary Material | Related Articles
    • Construction and Application of Knowledge Graph for Home Service Robot
    • YAO Rihui, CHEN Wenbai, CHEN Qili, WU Peiliang
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 72-78.
    • Abstract ( 421 )       
    • To improve the service robot's accurate acquisition and cognition of the semantic information of things in the home environment, a construction method of the knowledge graph of the home service robot is proposed, and the application test of the home service is carried out. First, a domain ontology suitable for the home environment is constructed according to the application requirements of home services, and a tructured method is used to represent the semantic attributes of objects and the knowledge of environment, users, and services in the home scenario, and store them in a relational database. On this basis, according to the domain ontology, the mapping and knowledge extraction are performed on the structured data of the knowledge base to complete the construction of the knowledge graph, and the application requirements of typical home services are tested. The experimental results show that the constructed knowledge graph can effectively help the service robot to accurately obtain the semantic information of things in the home environment and retrieve the knowledge required for the service so as to ensure the intelligent execution of the home service.
    • Supplementary Material | Related Articles
    • Network Traffic Prediction Using Wavelet Denoising and Optimized Support Vector Machine
    • TIAN Zhongda, PAN Xinpeng
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 79-84.
    • Abstract ( 237 )       
    • In order to improve the accuracy of network traffic prediction, a network traffic prediction model is proposed based on wavelet denoising and improved slime mold algorithm optimized support vector machine. First, wavelet denoising is used to denoise network traffic, and support vector machine is used as the prediction model. Since the prediction results of support vector machine are greatly affected by the model parameters, an improved slime mold algorithm with random inertia weight is used to optimize the penalty factor and kernel function parameters that used in the support vector machine model. The validity of the proposed model is verified by the collected network traffic. The simulation results show that the proposed model is superior to the comparison model in terms of the evaluation index.
    • Supplementary Material | Related Articles
    • Image Feature Extraction Algorithm Based on Orthogonal Projection Learning
    • ZHANG Xiaoqian, TAN Zhen, WANG Xiao, LIANG Qin, WAN Liming
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 85-90,128.
    • Abstract ( 275 )       
    • To overcome the deficiencies of low-rank embedding in data reconstruction and noise suppression, and improve the accuracy of its feature recognition,an image feature extraction algorithm is proposed based on orthogonal projection learning. The half-quadratic alternating direction method of multipliers algorithm is designed to solve the orthogonal projection learning model. The model retains the main features of the samples by introducing an orthogonal matrix,the norm constraints makes the extracted features more prominent, and the weighted Schatten p-Norm is used to approximate the optimal solution of the rank. To improve the robustness of the model and make it suitable for supervised scenarios, generalized correntropy is used for data item modeling and classification loss function construction. Experimental results on different scale datasets show that the proposed model has better feature extraction performance than other existing models.

    • Supplementary Material | Related Articles
    • Mapping-Varied Spatial Modulation for Secure Transmission Based on Multi-RIS Optimization
    • DING Qingfeng, YANG Qian, XU Mengyin
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 91-96.
    • Abstract ( 315 )       
    • In order to ensure the anti-eavesdropping security of signal transmission, a mapping-varied strategy based on multiple reconfigurable intelligent surfaces ( RIS ) is proposed. First, based on instantaneous channel quality between RIS and transmitter, the transmitter selects different mapping modes of spatial modulation and constellation modulation to improve the difficulty of eavesdropping. In order to ensure the security rate and bit error rate of legitimate receivers, the position and phase shift of RIS are optimized to maximize the signal-to-noise ratio of legitimate receivers. Meanwhile, the gradient projection method is used to solve the non-convex constant mode constraint problem of RIS phase shift. Simulation results show that the proposed secure transmission strategy can guarantee the performance of the legitimate receiver at a lower signal-to-noise ratio. Meanwhile, in instantaneous mapping mode, the eavesdropper cannot decode the eavesdropping information correctly, which greatly ensures the anti-eavesdropping security of the system.
    • Supplementary Material | Related Articles
    • An Improved Sensitivity Encoding Reconstruction Algorithm Based on Sparse Transform Learning
    • LI Xilan, DUAN Jizhong
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 97-102.
    • Abstract ( 249 )       
    • The sensitivity encoding ( SENSE) technique utilizes sensitivity information from multiple receiving coils to reduce scan time. The existing SENSE-based parallel MRI reconstruction methods have problems of artifacts and missing details, which is not conducive to clinical diagnosis. By introducing data-driven adaptive sparse transform learning (TL) into the SENSE algorithm, TL-SENSE algorithm is proposed, that reduce the artifacts and improve the quality of parallel MRI reconstruction. The proposed algorithm employs the alternating direction method of multipliers (ADMM) to solve the target optimization problem. And The proposed algorithm comprises three steps: transform updating, hard threshold denoising and image updating. The simulation results show that the proposed algorithm performs well in image denoising and restoration and preserves the texture details and edge information. It also achieves higher consistency between the reconstructed image and the original image. For the selected 48 sets of data, the average signal noise ratio of TL-SENSE increased by 4.62 dB, 1.91 dB, 1.30 dB and 0.89 dB compared with that of SENSE, L1-SENSE, TV-SENSE, and LpTV-SENSE, respectively.

    • Supplementary Material | Related Articles
    • Improved YOLOv3 Algorithm for Multi-Target Detection of Traffic
    • SONG Yubo, GAO Jiazhen
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 103-108.
    • Abstract ( 310 )       
    • A YOLOv3-based multi-target detection method is proposed to address the problem of missed and false detection caused by the small percentage of target pixels and mutual occlusion in traffic scenes. The method implants a spatial pyramid pooling module in the YOLOv3 network structure to enhance feature representation, and a multi-scale feature fusion mechanism is proposed to obtain both spatial information and semantic information. The semantic information of the target to be detected is refined by extending the prediction branch of the prediction layer. In addition, the improved K-means + + clustering algorithm is used to extract the initial center of the priori box and improve the matching degree between the prediction anchor box and the target to be detected. Meanwhile, a flexible non-maximum suppression algorithm is applied to adjust the confidence score flexibly. The experimental results based on the hybrid data set show that the proposed method improves the detection accuracy effectively.

    • Supplementary Material | Related Articles
    • Knowledge Driven Management Strategy of Human-Machine Active Dialogue
    • HUANG Hongcheng, KONG Tiantian, HU Min, TAO Yang, KOU Lan
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 109-114.
    • Abstract ( 285 )       
    • To solve the problem that the current dialogue system is mainly passive response and still unable to carry out active dialogue well, a knowledge driven human-machine active dialogue management strategy is proposed, which simulates human communication mode and divedes the dialogue into two sub-tasks: topic switching and topic depth. A personalized dialogue management strategy is designed to realize active guidance and topic transfer in multi-round dialogues. The proposed strategy determines the time of the system's active dialogue based on the emotional state of human-machine interaction, and uses the knowledge graph as the background knowledge information to actively search the multi-hop neighbor set of the dialogue entities that are triggered by the knowledge graph, so as to determine the next interaction content. For topics of users' negative emotions, new topics are actively sought for through outward communication method. For topics with users' positive emotions, the current topic can be deeply respond to through cohesion. The experimental results show that the initiative of model dialogue is improved by the strategy while balancing global dialogue coherence and local topic consistency, which is a new reference for the development of human-machine active dialogue system.
    • Supplementary Material | Related Articles
    • An Adaptive Bit Rate Algorithm Model Based on Wireless Network
    • CHEN Chunlei, LIU Kaijun, DONG Chen, ZHOU Hongyuan
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 115-120.
    • Abstract ( 263 )       
    • To solve the problem of drastically changing network fluctuation and long tail, an adaptive bit rate algorithm model based on deep learning is proposed. The algorithm optimizes video bit rate, bit rate switching frequency, and video pause time, and it is robust to the randomness of network fluctuations. Compared with other models, the proposed model has better performance in the scenario of drastic fluctuations in wireless networks. Under the worst network conditions, the comparison model causes video playback lag with a probability as high as 16% , while the proposed model has a lag probability of only 1% , and the average quality of experience index is 30% higher than that of the comparative model.
    • Supplementary Material | Related Articles
    • Trajectory Estimation Algorithm for Unmanned Aerial Vehicle Based on LSTM-KF
    • LIU Jinming, ZHANG Yuyan, ZHANG Biling
    • Journal of Beijing University of Posts and Telecommunications. 2022, 45(5): 121-128.
    • Abstract ( 800 )       
    • In the case of limited measurement information, Kalman filter (KF) is difficult to deal with unmanned aerial vehicle tracking by using a single motion model. To solve this problem, a novel long short-term memory(LSTM)-KF algorithm combining LSTM and KF algorithm is proposed. First, LSTM is used to predict the average and instantaneous velocity of the target so that the problem of poor generalization ability of nonparametric model can be solved in position prediction task. Then, the prediction limitation of KF algorithm using motion model is analyzed, and the method of using LSTM prediction results to modify the prediction results of motion model is proposed to reduce the prediction error. The revised prediction results are combined with the measurement data to realize the state estimation of the target. Finally, the proposed algorithm is verified on the generated trajectory. The simulation results show that LSTM-KF algorithm has higher tracking accuracy and stronger robustness than the existing models.

    • Supplementary Material | Related Articles