Please wait a minute...

Journal of Beijing University of Posts and Telecommunications

  • EI核心期刊

Current Issue

    • 2025, Vol. 48 No. 1 Published date:26 February 2025 Last issue
    • DDPG-Based Access Point Selection in Cell-free Massive MIMO System
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 1-6,13.
    • Abstract ( 153 )       
    • To avoid the overload of fronthaul links in cell-free massive multiple input multiple output (MIMO) system, an optimization problem for access point selection is formulated to maximize the fronthaul link usage efficiency. Firstly, the average spectral efficiency per-link is constructed to measure the utilization efficiency of the fronthaul link. Secondly, the objective problem of maximizing the average spectral efficiency per-link while meeting users’ quality of service requirements is established. Then, considering the non-convexity of the objective problem and the non-differentiability of the optimization variables, which makes it challenging for traditional optimization algorithms to find solutions, a dynamic access point selection algorithm based on deep deterministic policy gradient ( DDPG ) is proposed. Finally, the output of the DDPG is refined to ensure compatibility with the discrete nature of access point selection actions. Simulation results show that the proposed algorithm can effectively enhance the average spectral efficiency per link of the system and meet the quality of service requirements.
    • Supplementary Material | Related Articles
    • A PAPR Reduction SLM Method for OTFS Signals without Transmitting Side Information
    • XU Chen, WANG Hao, SHAO Siyu, REN Guangliang
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 7-13.
    • Abstract ( 137 )       
    • Orthogonal time frequency space ( OTFS) modulation has an excellent performance in high-speed mobile scenarios. However, OTFS modulation is susceptible to the problem of high peak-to-average power ratio (PAPR). Regarding this issue, a selective mapping (SLM)method without transmitting side information is proposed to reduce the PAPR of OTFS signals. The random phase sequences are employed to scramble the transmitted symbols in the delay-Doppler ( DD) domain, and the unique degree of freedom of OTFS pulse pilot positions are used to carry the side information, which can be extracted by detecting the differences in the energy distribution of received symbols in the DD domain. Finally, the input-output relationship in the DD domain was derived again, and an improved channel estimation and signal detection algorithm was provided. Simulation results show that the proposed method can effectively reduce the PAPR by about 2.4 dB, and the performance loss of bit error rate is only 0.4 dB with the estimated channel state information.
    • Supplementary Material | Related Articles
    • Two-Stage Approach Based on Multi-Semantic Representation for Interactive Argument Pair Extraction
    • YU Ning, SHI Yu, LIU Jianyi
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 14-20.
    • Abstract ( 21 )       
    • Interactive argument pair extraction aims to identify multiple argument pairs from two related argumentative texts. The task is decomposed into two subtasks by current methods: argument mining and argument pair extraction, which are then jointly modeled for multi-task learning. However, the same sentence semantic representations are shared by these methods, which may lead to information bias within each subtask. Therefore, a two-stage interactive argument pair extraction method based on multi-semantic representation is proposed. First, a pre-trained language model and a bidirectional encoder are utilized in the argument mining stage to capture the semantic information of a single text for argument identification. Then, the two arguments are directly modeled as a text pair for semantic representation in the argument pair extraction stage. A large language model is used to extract key information from lengthy argument content, new samples are generated, key semantic information is introduced, and ultimately the identification of interactive argument pairs is completed. Experiments are conducted on two benchmark datasets, the results show that the proposed method outperformed the baseline method, demonstrating its effectiveness.
    • Supplementary Material | Related Articles
    • Adaptive Density Peak Clustering Algorithm Based on Comparative Parameters
    • SHAO Zhuang, FU Weihong
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 21-25.
    • Abstract ( 91 )       
    •  In response to the shortcomings of the density peak clustering algorithm, a novel algorithm named adaptive density peak clustering based on comparative parameters is proposed. In the proposed algorithm, a new metric named relative local density was used to assess the similarity between clusters, which greatly improved its applicability to datasets. Another variable called relative connectivity distance was applied for measuring the similarity between clusters, which effectively eliminates the influence of different sizes of clusters in the dataset. The applicability of the algorithm on different datasets was enhanced. By constructing a comentropy function, the parameters could be adaptively determined according to the characteristics of the datasets, which improved the intelligence of the algorithm. A new point allocation strategy was proposed to avoid the effects of the ‘chain reaction’. Experimental results show that the proposed algorithm significantly improves clustering performance compared to the standard and improved density peak clustering algorithm.
    • Supplementary Material | Related Articles
    • Relay-Enabled Backscatter Communication Networks: Hybrid Transmission Scheme’s Design and Optimization
    • XU Rui, YE Yinghui, LU Guangyue
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 26-32.
    • Abstract ( 473 )       
    • For the relay-enabled backscatter communication network, the entire transmission time block is divided into the backscatter phase and the symmetric transmission scheme based relay-enabled backscatter phase, and a hybrid transmission scheme is proposed, based on which a throughput maximization resource allocation scheme is designed. Specifically, a multidimensional resource allocation problem is formulated to maximize the throughput by jointly optimizing the time allocation factor, the power reflection coefficients in each phase of the Internet of things node and the transmitting power in each phase of the hybrid access point. As the formulated optimization problem includes couplings among optimization variables and the max-min function, it is non-convex and can not be solved by the existing convex tools. Therefore, a series of slack variables and auxiliary variables are introduced and the situational discussion is applied to convert the formulated problem into a convex one. Finally, computer simulations validate the superiority of the proposed scheme.
    • Supplementary Material | Related Articles
    • Deep Analysis and Detection of Anomalous Data Based on Dual-Layer Attention
    • GUO Gaoqiang, HE Mingshu, LI Xinhang, WANG Xiaojuan
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 33-38.
    • Abstract ( 115 )       
    • The development of intelligent driving technology increased vehicles interactions with external networks. The controller area network is the primary in-vehicle network and its security vulnerabilities can be exploited by attackers to gain control of vehicles, posing significant safety threats to occupants. To address this issue, an anomaly detection method based on long short-term memory and an attention mechanism is proposed. A dual-layer attention encoder is employed to deeply extract information from both local and global features within the data flow of the controller area network. By efficiently identifying and learning sequential patterns among features, the proposed method requires significantly less feature information, achieves better detection efficiency and accuracy, and enables unbalanced classification tasks. Finally, multiple independent repeated tests are performed on the CarHacking dataset. The results demonstrate that detection accuracy exceeding 99.20% has been achieved for all attack categories in the dataset, significantly outperforming existing detection methods. Additionally, a multi-classification task that could not be achieved by other approaches is accomplished by the proposed model, with an average detection accuracy of 99.26% .
    • Supplementary Material | Related Articles
    • Opportunistic Network Link Prediction Based on Temporal Generative Adversarial Networks
    • SHU Jian, WANG Pengtao, LI Ruirui
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 39-45.
    • Abstract ( 98 )       
    • The frequent node mobility of opportunistic networks poses significant challenges for link prediction. To better reflect the temporal evolution of the topology, an opportunistic network link prediction method based on temporal generative adversarial networks is proposed. A network volatility is defined for calculating the network segmentation duration so that an opportunistic network is sliced into fine-grained snapshots. Information matrices are extracted from these network slices which integrate spatial and temporal dimensional information, and network feature vector matrices are constructed by graph embedding methods. Combining gated recurrent units and generative adversarial network, a temporal generative adversarial network model is developed. It learns the evolutionary features of network topology, so as to achieve link prediction in networks for the future. The experimental results on three real datasets demonstrate that the predictive performance of the proposed method is superior to that of the baseline method.
    • Supplementary Material | Related Articles
    • Multi-Party Human-Computer Active Dialogue Strategy Based on Knowledge Enhancement
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 46-51.
    • Abstract ( 61 )       
    • In view of the problem that the existing multi-party human-computer dialogue system is prone to ignore speakers who may be forgotten during the dialogue process, leading to weak user interaction initiative and poor user dialogue experience,a multi-party human-computer active dialogue strategy based on knowledge enhancement is proposed. The strategy uses a knowledge graph as external knowledge to learn the preferences of specific individuals who may be overlooked in the dialogue, combine the knowledge with the current group dialogue requirements to design a personalized response strategy, thereby encouraging individual active participation in group conversations. Initially, a graph attention mechanism is employed to learn the representation of interest entities specific to individuals, and a deep preference representation for individuals is obtained by introducing time-weighted aggregation. Subsequently, multiple neighbor sets are triggered along the path of the current dialogue topic knowledge subgraph to actively capture the high-level personalized interest topics of individuals. Finally, through semantic-level analysis of the current group dialogue context and requirements, the satisfaction of the group with candidate topics is evaluated, and the optimal dialogue topic is generated, which both satisfy individual preferences and consider group sentiments. Experimental results indicate that a multi-party human-computer dialogue system, which integrates external knowledge and focuses on specific speakers, effectively enhances the richness of response content and interaction satisfaction from participants, fostering continuous multi-party dialogue.
    • Supplementary Material | Related Articles
    • Evaluation of Node Importance in Opportunistic Networks Based on Node Embedding
    • LIU Linlan, CUI Hui, GAO Haoxuan, SHU Jian, JIANG Yunan
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 52-58.
    • Abstract ( 70 )       
    • To improve the accuracy of node importance evaluation in opportunistic networks, a node embedding-based node importance evaluation method is proposed. Considering the time-varying nature of opportunistic networks, a time window aggregation graph is employed, so that network topology and temporal connection data can be obtained in the window. Graph attention mechanism is utilized to extract the topological features of nodes, and temporal encoding and self-attention mechanism are employed to capture temporal features of nodes. Node embedding vectors are achieved by integrating two features. The cluster importance for nodes is introduced and the transition probability matrix is constructed. The node importance is obtained by PageRank algorithm. On real opportunistic network datasets, experimental results demonstrate that the proposed method exhibits better evaluation accuracy compared to approaches such as f-PageRank and dynamic graph convolutional network.
    • Supplementary Material | Related Articles
    • Fast Clustering Adaptive Rate Compressive Sensing for Surveillance Videos Subblocks
    • WANG Jianming, LUO Ping, YANG Qingqing, PENG Yi
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 59-65.
    • Abstract ( 77 )       
    • The distributed characteristics of video surveillance systems are evident, and the complexity of the sampling side can be reduced by adopting an appropriate distributed video sampling scheme. Since compressive sensing also exhibits distributed characteristics, it is considered to have great application potential in distributed video sampling schemes. In order to facilitate the application of compressive sensing in distributed surveillance video systems, adaptive rate compressive sensing is investigated. Based on statistical characteristics estimation, the mean and variance of unknown original video subblocks are estimated from the measurements obtained through compressive sensing, and subblocks are rapidly clustered. The sparsity of subblocks in each cluster is then estimated, and the sampling rate is allocated accordingly. The video adaptive rate compressive sensing is implemented, and the consumption of sampling rate is reduced without causing degradation in video reconstruction quality or significantly increasing sampling complexity. Simulation results demonstrate that the proposed method effectively allocates the sampling rate, and the computational complexity of sampling meets the requirements of actual sampling equipment.
    • Supplementary Material | Related Articles
    • A Method to Maximize Energy Efficiency in STAR-RIS Assisted Uplink NOMA Systems
    • TIAN Xinji, MENG Haoran
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 66-72.
    • Abstract ( 86 )       
    • A method to maximize energy efficiency is proposed for uplink non-orthogonal multiple access (NOMA) systems assisted by simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS). Firstly, the optimization problem of maximizing energy efficiency is constructed, with base station beamforming, STAR-RIS phase shift, users’ power and time allocation as variables. Then, the problem is simplified by deducing the relationship between beamforming and STAR-RIS phase shift as well as the minimum power required by the users. Finally, an iterative method is proposed to optimize STAR-RIS phase shift, power allocation and time allocation alternately. In each iteration, semi-definite programming method is used to optimize STAR-RIS phase shift, Dinkelbach algorithm to optimize power allocation, and function extremal method to optimize time allocation. The simulation results show that in the time switching protocol, time allocation has little effect on energy efficiency, while STAR-RIS phase shift has great effect on energy efficiency.
    • Supplementary Material | Related Articles
    • Field Road Defect Detection Method Based on Edge Intelligence
    • CHEN Zeyu, GU Yue, CHENG Siyao, FENG Guohui
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 73-78.
    • Abstract ( 85 )       
    •  In the autonomous operation tasks of smart agricultural machinery, real-time field road defect detection is considered essential to ensure the safe operation of the machinery. However, existing road defect detection technologies have been minimally explored in the agricultural domain, and a dedicated dataset for field roads has yet to be developed. To obtain a defect detection model specific to field roads, a model was initially trained on a conventional road pothole dataset, and then followed by transfer learning using a simulated field road pothole dataset. To address the issue of reduced detection accuracy after transfer learning, an attention mechanism was introduced into the YOLOv5s network architecture to enhance the network’s precision, achieving an accuracy of 83.15% for defect detection in field road scenarios and satisfying its accuracy requirements. To verify the edge performance of the defect detection model, it was deployed onto the Jetson Nano for simulation experiments. To meet the real-time detection requirements of the field road defect detection model on edge devices, TensorRT was used to optimize and compress the model, and the pothole detection speed was improved from 396 milliseconds per frame to 157 milliseconds per frame.
    • Supplementary Material | Related Articles
    • Minimum User Rate Maximization Optimization Algorithm for IRS Assisted Wireless Powered Communication Network
    • LEI Weijia, ZHANG Tong, LEI Hongjiang, TANG Hong
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 79-86.
    • Abstract ( 86 )       
    • To improve the overall system transmission rate while ensuring fairness among users in intelligent reflecting surface ( IRS) assisted wireless powered communication network, an optimization algorithm is proposed that jointly optimizes hybrid access point beamforming, uplink-downlink time allocation, user transmission power, and IRS phase shift matrix, with the objective of maximizing the minimum user transmission rate. The original non-convex optimization problem with coupled variables is transformed using relaxation variables and semi-definite relaxation. Methods such as golden section search, successive convex approximation, and alternating iterative optimization are then applied to solve it. Simulation results indicate that, compared with the algorithm aiming to maximize the sum rate, the proposed algorithm significantly improves the minimum user rate and enhances fairness among users. Additionally, compared to the algorithm without IRS, the proposed algorithm achieves higher minimum user rates and system sum rates, with performance improving as the number of IRS elements increases, indicating that IRS effectively enhances both energy and information transmission efficiency.
    • Supplementary Material | Related Articles
    • Blockchain Ciphertext Data Sharing and Access Control Scheme Based on Zero-Knowledge Proof
    • REN Zhixin, YAN Enhua, CHEN Taowei, YU Yimin
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 87-91,113.
    • Abstract ( 206 )       
    • The majority of existing solutions are observed to still retain an authorization authority on the blockchain, which is viewed as exacerbating the “blockchain trilemma” issue. To address this, a zero- knowledge proof mechanism has been introduced in order to break away from the traditional model where key management and distribution are conducted by a central authority or trusted third party. Initially, a re-encryption protocol is employed to enable secure management and distribution of the attribute-based encryption algorithm’s master key without the involvement of an authorization authority. The correctness of the off-chain computations in re-encryption is then verified by using zero-knowledge proofs. Finally, a transaction aggregation circuit for blockchain ciphertext access control is developed to enhance system scalability and reduce on-chain costs. Through theoretical analysis and simulation, it is demonstrated that, in comparison to the traditional key management and distribution model, secure and efficient data sharing and access control are achieved, and on-chain overhead is effectively reduced.
    • Supplementary Material | Related Articles
    • The Low Latency Routing Solution for Burst Services in Optical Transport Network
    • ZHOU Pengfei, YUAN Chao, FU Zhenxiao, WANG Yimeng, ZHAO Yongli
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 92-99.
    • Abstract ( 60 )       
    • To solve the problem of increased network service blocking rate and average latency caused by burst service, a burst traffic problem model in optical transport network was established. By adjusting the maximum bandwidth occupancy rate, the blocking rate in burst and non-burst states is balanced. The model also sets an optimization objective function to reduce the burst blocking rate and the average latency. A joint optimization algorithm based on latency and maximum bandwidth occupancy rate is proposed in the model. This algorithm is based on the K-shortest paths (KSP) algorithm with remaining link bandwidth and transmission latency as constraints and optimization objectives. Among the candidate paths obtained, the path with the minimum optimization objective function established in the proposed model and sufficient bandwidth resources is selected to transmit service. The experimental results show that compared with other reserved bandwidth algorithms, the proposed algorithm has the lowest traffic blocking rate, up to 18.16% reduction compared to the reserved bandwidth-based multipath transmission algorithm. Moreover, the average service latency has always remained low up to 0.36 ms reduction compared to the reserved bandwidth-based multipath transmission algorithm. The proposed algorithm achieves optimization of transmission latency and blocking rate for burst services.
    • Supplementary Material | Related Articles
    • Phoneme Recognition Based on B-Wave-U-Net Feature Enhancement at Low Signal-to-Noise Ratio
    • HUANG Huibo, SHAO Yubin, LONG Hua, DU Qingzhi
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 100-106.
    • Abstract ( 74 )       
    • To address the issue of low phoneme recognition accuracy at low signal-to-noise ratios (SNR), a phoneme recognition method is proposed based on B-Wave-U-Net feature enhancement. First, a bidirectional long short-term memory (BLSTM) network is integrated at the beginning side of the Wave- U-Net encoder, from where the information flow is extracted and jump-connected to the decoder side. Then it will be inserted into a fully connected layer to form the B-Wave-U-Net network. The next speech spectrogram is then enhanced and denoised using the B-Wave-U-Net. Finally, Mel filtering is applied to extract the log-Mel scale bank energy features. Phoneme recognition tests are conducted under 0 dB SNR with a white noise source, using the THCHS30 dataset and the ResNet-BLSTM-CTC model. Experimental results show that the proposed B-Wave-U-Net outperforms the baseline network, reducing the phoneme error rate by 0.9% to 2.5% . This demonstrates the significant advantage of the B-Wave-U-Net in robust feature extraction for phoneme recognition under noisy conditions.
    • Supplementary Material | Related Articles
    • Channel Information-Dynamic Weighted Based Decoding Algorithm for Non-Binary LDPC Codes
    • CHEN Haiqiang, XIAN Wenluo, LI Yulin, LI Qingnian, LI Xiangcheng
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 107-113.
    • Abstract ( 72 )       
    • A dynamic channel information-weighted based decoding algorithm for non-binary low density parity check (LDPC) codes is proposed, which can dynamically weight the decoding reliability according to channel information. Firstly, the complicated real-number multiplication operation in reliablity factors is replaced by an additive operation, thus can reduce the computational complexity. Secondly, the reliability objective function is redesigned based on the initial channel information, which can enhance the stability of the iterative decoding information and boost decoding performance. Furthermore, a weighted coefficient is introduced, which can be dynamically changed with respect to the signal-to-noise ratio. The reliability information can be dynamically revised and updated in the decoding iterations to improve the decoding performance. Simulation results show that the proposed algorithm outperforms the recently presented adaptive single / multiple distance symbol flipping decoding algorithm with prediction.
    • Supplementary Material | Related Articles
    • Active Reconfigurable Intelligent Surface Topology Optimization and Self-interference Elimination Scheme
    • TANG Jinmin, TAO Mei, SONG Yaolian, YU Guicai
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 114-119.
    • Abstract ( 120 )       
    • In order to suppress the self-interference between active reconfigurable intelligent surface (RIS) components and further improve the system capacity, a joint optimization scheme for active RIS topology and precoding design is proposed. Under the maximum power constraints of base station and active RIS, the optimization problem is established based on the criterion of maximum weighted sum-rate (WSR). In order to solve the non-convex problem, the problem is decomposed into two subproblems, active RIS topology and precoding design. The subproblems are transformed and iteratively optimized by using adaptive tabu search, Lagrange multiplier method and binary search algorithm, and an approximate suboptimal solution for the non-convex problem is obtained. The simulation results show that the WSR of the proposed scheme is 3.75 bit·s-1·Hz-1 higher than that of the conventional scheme under ideal conditions. In the case of self-interference, the WSR of the proposed scheme only decreases by 0.46% compared with that of the ideal scheme versus 1.6% drop in the conventional scheme, demonstrating that the proposed scheme can effectively mitigate the self-interference between active elements.
    • Supplementary Material | Related Articles
    • The Age of Information Performance Analysis in NOMA-MEC Status Update Systems
    • LIU Lei, QIANG Jingzhou, ZHANG Xuewei, JIANG Fan, WANG Junxuan
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 120-126.
    • Abstract ( 268 )       
    • In order to characterize the information freshness of the mobile edge computing (MEC) state update system based on non-orthogonal multiple access (NOMA), the information age (AoI) of users in two scenarios was studied. First queuing theory is utilized to mimic the transmission and computation processes of status updates information. After analysis of the transmission queue’s service rate that hinges on the quality of NOMA transmission, closed-form average AoI is derived for local and edge computing schemes. The validity of the theoretical analysis results is verified through Monte Carlo simulations. The simulation results that, compared to orthogonal multiple access, the system AoI performance can be enhanced more dramatically by applying NOMA-MEC with a higher server computation rate. In addition, the effects of the system parameters, such as server computation rate, status update information size, and transmission power, on the AoI performance are demonstrated by the simulation results.
    • Supplementary Material | Related Articles
    • Neural Network Ensemble Models for Financial Time Series Forecasting
    • ZHANG Han, WANG Weiguo
    • Journal of Beijing University of Posts and Telecommunications. 2025, 48(1): 127-132.
    • Abstract ( 168 )       
    •  The accurate forecasting of financial time series data holds a significant position in the operation and management of financial markets. Grounded in the notions of neural networks and ensemble learning, the convolutional neural network ( CNN), long short-term memory ( LSTM) network, and autoregressive moving average ( ARMA) model are integrated within an ensemble framework to put forward a novel ARMA-CNN-LSTM model for forecasting financial time series data. The spatiotemporal characteristics within the data are modeled by the CNN-LSTM model, while the autocorrelation features of the data are simultaneously modeled by the ARMA model, achieving hybrid modeling of linear and nonlinear traits in financial time series data. The experimental results show that, compared with the baseline individual models, the proposed model demonstrates excellent performance in both the accuracy and robustness on predicting financial time series data.
    • Supplementary Material | Related Articles