Top Read Articles

Published in last 1 year |

In last 2 years |

In last 3 years |

All

Please wait a minute...


For Selected: Download Citations EndNote Reference Manager ProCite BibTeX RefWorks Toggle Thumbnails

Select

Improved YOLOv5 for ship detection in SAR images with arbitrary orientations QU Chunhui, WANG Wei, ZHANG Ting, WANG Yinghua, CHEN Bo Journal of Xidian University    2025, 52 (4): 1-14.   DOI: 10.19665/j.issn1001-2400.20250311 Abstract （817）   HTML （132）    PDF（pc） （3180KB）（513）       Knowledge map    Save The detection of ships in Synthetic Aperture Radar (SAR) images faces challenges such as multi-scale,arbitrary orientations,and dense arrangements.While rotated bounding box detection methods can accurately identify ships in any orientation,existing approaches struggle to balance high accuracy with real-time performance.To address these issues,we propose a novel rotated bounding box detection network model that combines midpoint offset representation and YOLOv5.This model inherits the regression mechanism of the horizontal bounding box and introduces a multi-task joint loss function suitable for rotated detection,thus overcoming the challenges of training instability and parameter redundancy caused by angle-based representations in conventional rotated box methods.Additionally,we propose a backbone network based on attention mechanisms,which integrates both global and local image features to enhance the importance of regions of interest and improve the model’s feature extraction capabilities.The proposed lightweight rotated bounding box detection network demonstrates an improved performance.Experiments on the RSDD-SAR dataset show that our method effectively detects ships in SAR images at any orientation and provides corresponding angle information.Compared to existing rotated box detection methods,our approach achieves superior detection results with an average precision of 90.02%.In particular,in complex nearshore scenarios,the proposed method outperforms others,reaching an optimal average precision of 70.5%,while the model has a compact parameter size of only 7.2 M.Additional experiments further validate the effectiveness of the proposed method. Table and Figures | Reference | Related Articles | Metrics

Select

Research on system loganomaly detection based on semi-supervised learning SUN Fuxiong, MENG Wenjin, SUN Yiling, LI Yangyi, YIN Yukai Journal of Xidian University    2025, 52 (4): 15-32.   DOI: 10.19665/j.issn1001-2400.20250312 Abstract （442）   HTML （50）    PDF（pc） （2397KB）（147）       Knowledge map    Save Distributed system logs are generated by components or nodes within distributed systems,recording time-sequenced data of system operational statuses and events.They are characterized by diversity,large volume,poor readability,and security vulnerabilities.Current log anomaly detection methods face dual limitations:(1) supervised approaches require expert manual labeling when log annotations are insufficient,which is costly; (2) unsupervised methods struggle to construct effective detection models.To solve these problems,this paper proposes a semi-supervised log anomaly detection model.First,log data are transformed into event template sequences.The unitary invariance of word embeddings and matrix perturbation theory are employed to determine optimal word vector dimensions,followed by constructing temporal log training sets.Second,multiple deep learning architectures are investigated,including the Temporal Convolutional Network with Attention (TCNA),Bidirectional LSTM (BiLSTM),and other network structures.Experiments demonstrate that under a 96.98% data imbalance,all supervised models achieve a good performance,with the TCNA exhibiting the best results.Finally,the MixMatch and FixMatch semi-supervised algorithms are integrated into the TCNA,thus building the SSLogTCNA model.The loss function and data augmentation strategies are optimized for log sequence characteristics.Experimental results indicate accurate predictions on hundreds of thousands of unlabeled samples using minimal labeled data.Comparative experiments show the SSLogTCNA with the MixMatch achieves an optimal performance.Incremental experiments further verify the SSLogTCNA’s effectiveness:with only 0.1% labeled data,it approaches a fully-supervised detection performance.This work provides a novel solution to insufficient log labels and high annotation costs. Table and Figures | Reference | Related Articles | Metrics

Select

Jointcolor fusion and feature enhancement network for low-light image dehazing WANG Keyan, ZONG Xingpeng, CHENG Jicong, DONG Xinyu, LI Yunsong Journal of Xidian University    2025, 52 (5): 1-12.   DOI: 10.19665/j.issn1001-2400.20250601 Abstract （412）   HTML （68）    PDF（pc） （3892KB）（275）       Knowledge map    Save Existing dehazing networks exhibit a limited capability in feature extraction and color bias suppression under low-light haze conditions, often leading to detail loss and color distortion. To address these issues, we propose FCformer, a Joint Feature Enhancement and Color Fusion Network for low-light image dehazing. To restore the image structure and texture, a feature enhancement backbone is designed with window-spatial and sparse-channel modules to focus on key local and global features. A color fusion branch, by incorporating color correction and fusion, improves chromatic representation. A learnable prior constraint module based on atmospheric scattering and Retinex models regularizes the output. Finally, a composite loss function, by combining reconstruction, perceptual, and color losses, guides better detail and color restoration. Experiments show that the FCformer surpasses the DehazeFormer by 0.98 dB in PSNR with a similar parameter size, and achieves a PSNR comparable to that of the ACANet while reducing parameters by 96.84%, demonstrating a superior visual performance. Table and Figures | Reference | Related Articles | Metrics

Select

Spatial-temporal correlation-based anomaly detection in UAV flight data GU Zhaojun, WANG Jingyu, WANG Jialiang, NIE Liuyang Journal of Xidian University    2025, 52 (4): 33-45.   DOI: 10.19665/j.issn1001-2400.20250403 Abstract （411）   HTML （28）    PDF（pc） （3233KB）（132）       Knowledge map    Save To address the limitations of current UAV anomaly detection methods,including suboptimal feature selection from flight data,inadequate spatio-temporal modeling,and low discriminative power of decision criteria,this paper proposes a data-driven UAV anomaly detection model named UAV-STAD which enhances anomaly detection by optimizing feature selection,integrating spatiotemporal features,and refining anomaly criteria.First,irrelevant features are filtered out using the Maximal Information Coefficient (MIC),ensuring that only significant features are retained for analysis.Second,relevant features are modeled as nodes in a graph,where an Adaptive Inter-Variable Correlation Learning (AIVCL) module computes variable adjacency matrices to facilitate spatial modeling with Graph Attention Networks (GAN).For temporal modeling,learnable Gaussian kernels are incorporated into the Transformer architecture to work in parallel with the self-attention mechanism by computing prior and sequential dependencies respectively.Finally,the model quantifies the differences in associations derived from both prior and sequential dependencies,combining these differences with reconstruction errors to define the anomaly criteria.The performance of the UAV-STAD is evaluated on the Thor flight datasets 97,ALFA and UAVSet,achieving average F1 scores of 97.2%、96.6% and 95.7%. Table and Figures | Reference | Related Articles | Metrics

Select

Lightweight self-supervised monocular depth estimation combining multi-scale attention GE Jingrui, QIN Guoxuan, ZHANG Wei Journal of Xidian University    2025, 52 (4): 66-76.   DOI: 10.19665/j.issn1001-2400.20250310 Abstract （320）   HTML （10）    PDF（pc） （1713KB）（59）       Knowledge map    Save To address the issues of large parameters,high computational complexity and difficulties in deploying models on edge computing devices for real-time inference in current monocular depth estimation networks,a lightweight self-supervised monocular depth estimation method with multi-scale attention is proposed which introduces a multi-scale attention module as the main body of the encoder,with the core concept centered on local structure information from convolution operations and long-range global information from self-attention mechanism,combining a gating multilayer perceptron with large kernel dilated convolutions and a feedforward neural network to achieve local and global feature aggregation with attention mechanisms,and to reduce parameters and computational complexity while ensuring the accuracy of depth estimation.The depth map obtained through the encoder-decoder architecture is used to reconstruct images together with the relative pose matrix output by a pose estimation network based on ResNet18.By calculating the photometric loss and smoothness loss between the reconstructed and original images,self-supervised monocular depth estimation is achieved.The algorithm has only 4.1M parameters and 3.0G FLOPs,with an average network structure runtime of 5.7ms.On the public KITTI dataset,the algorithm reaches 0.104 on AbsRel and 0.892 on,and its overall performance surpasses current state-of-the-art methods.Experimental results show that the algorithm achieves a high depth estimation accuracy and a fast inference speed,meeting the requirements for real-time monocular depth estimation tasks. Table and Figures | Reference | Related Articles | Metrics

Select

Vision transformer model compression based on pruning-distillation ZHENG Yang, JIANG Xiaotian, FU Donghao, GUO Kaitai, LIANG Jimin Journal of Xidian University    2025, 52 (4): 55-65.   DOI: 10.19665/j.issn1001-2400.20250104 Abstract （304）   HTML （14）    PDF（pc） （1678KB）（56）       Knowledge map    Save Currently,the Vision Transformer has demonstrated outstanding performance across various tasks in the field of computer vision.However,their complex network structures typically require substantial storage and computational resources,making widespread deployment on resource-constrained devices challenging.To address this issue,we propose a compression method for the Vision Transformer based on pruning and distillation,aiming to reduce the model size while ensuring performance retention.First,through a structural analysis of the Vision Transformer,we identify the targets for width pruning as the attention heads in the multi-head self-attention mechanism and the neurons in the hidden layers of the multi-layer perceptron.We then employ a parameter importance evaluation strategy based on changes in the model’s loss function to assess these parameters.Next,we apply a post-pruning distillation strategy to prune the model in the terms of width and restore the accuracy of the pruned subnetworks.Finally,in the depth dimension,we obtain the final compressed model through post-pruning distillation.The proposed method is experimentally validated on the Tiny ImageNet,CIFAR-100,and CIFAR-10 datasets,compressing the Vision Transformer.When reducing the parameter count and computational load by 30%,on the Tiny ImageNet dataset,the accuracy of the ViT-S model is decreased by only 0.3%,while the accuracy of the ViT-B model is even improved by 0.6%.Experimental results indicate that our proposed method effectively balances the model accuracy and compression ratio. Table and Figures | Reference | Related Articles | Metrics

Select

Resource allocation scheme for URLLC in the NR-V2X system YANG Yijin, CHEN Jian, ZHOU Yuchen, YANG Long, HE Bingtao Journal of Xidian University    2025, 52 (4): 106-119.   DOI: 10.19665/j.issn1001-2400.20230702 Abstract （279）   HTML （2）    PDF（pc） （1625KB）（35）       Knowledge map    Save The diversity of services in quality demanded by vehicles in new radio vehicle-to-everything systems can lead to conflicts in resource scheduling,complex interference and data congestion issues,thereby reducing the efficiency of vehicle data transmission.From the perspective of the overall system performance,a dual-scale resource optimization scheme is designed to ensure the long-term stability of the new radio vehicle-to-everything system.This scheme reduces the communication overhead caused by high-speed vehicle mobility by clustering vehicle multicast clusters.Additionally,it ensures low latency and high-reliability communication by imposing constraints on the age of information and signal-to-noise ratio probability.And the scheme optimizes the vehicle data arrival rate,system resource blocks,and transmission power of each vehicle user to achieve a maximum average total throughput of the system’s vehicle-to-infrastructure communication.To solve the mixed-integer non-convex optimization problem,the long-term optimization problem is decoupled into a series of instantaneous optimization problems using the Lyapunov optimization method.The optimal solution is obtained using bipartite matching algorithms and the optimization theory.Simulation results show that the proposed scheme can ensure millisecond-level communication latency and 99.9% link reliability for vehicle-to-vehicle multicast communication.Moreover,it can balance the optimal average throughput of vehicle-to-infrastructure communication and a stable cache queue by adjusting control parameters,achieving more efficient and adaptive resource management and communication services in practical applications. Table and Figures | Reference | Related Articles | Metrics

Select

Collusion-resistant multi-authority attribute-based encryption on lattices ZHAO Zongqu, MA Shaohua, GUO Menghao, WANG Naifeng Journal of Xidian University    2025, 52 (4): 165-179.   DOI: 10.19665/j.issn1001-2400.20250309 Abstract （273）   HTML （5）    PDF（pc） （1413KB）（35）       Knowledge map    Save Attribute-Based Encryption (ABE) provides flexible access control,but keys are generated and distributed solely by a central authority,leading to an excessive workload and increased vulnerability.If the master key is compromised,it poses significant security risks.To address this issue,Multi-Authority Attribute-Based Encryption (MA-ABE) allows multiple authorities to independently distribute keys based on the attributes they manage.This ensures that even if one key is compromised,the system remains secure.Like traditional ABE,MA-ABE still faces the problem of collusion by arbitrary unauthorized users,and even some authorities may be corrupted and collude with the adversary.To solve these challenges,a new lattice-based MA-ABE scheme is proposed that resists arbitrary collusion through randomization via tensors and avoids the exponential growth of noise caused by other pseudorandom functions or linear secret sharing schemes.The scheme is based on Evasive Learning with errors (Evasive LWE),removing the dependence on random oracles in security proofs and achieving static security without lattice pseudorandom functions or other non-standard assumptions.The scheme has a polynomial modulus,which improves the efficiency of the system modulus operations,makes the ciphertext more compact,and reduces the communication overhead. Table and Figures | Reference | Related Articles | Metrics

Select

Particle swarm optimization for trajectory deception jamming against networked radar systems WANG Xiaohan, LIU Congfeng Journal of Xidian University    2025, 52 (4): 180-191.   DOI: 10.19665/j.issn1001-2400.20250703 Abstract （254）   HTML （1）    PDF（pc） （2043KB）（35）       Knowledge map    Save In the complex and dynamic modern battlefield environment,traditional single-radar systems have become inadequate against contemporary warfare challenges,leading to the emergence of Networked Radar (NR) systems.By leveraging a unique common-source detection mechanism,NR systems exhibit superior detection and anti-jamming capabilities,rendering conventional jamming methods ineffective.This study focuses on Unmanned Aerial Vehicle (UAV) deception jamming techniques against NR systems,specifically proposing a trajectory deception jamming method based on an improved Particle Swarm Optimization (PSO) algorithm.First,the principle of multi-UAV cooperative trajectory deception jamming against NR systems is analyzed,with an optimization model for cooperative deception jamming established.Subsequently,the classical PSO algorithm is enhanced by introducing an adaptive Lévy flight hybrid PSO algorithm,which incorporates chaotic mapping,adaptive weight factors,Lévy flight,and differential evolution operations.This algorithm solves the cooperative deception jamming optimization model under multiple constraints,including kinematic feasibility and common-source detection consistency.Finally,simulation experiments verify the feasibility and effectiveness of the proposed method.Compared to existing algorithms,the proposed algorithm demonstrates significantly a higher computational efficiency and superior optimization precision,showcasing substantial potential for engineering applications in electronic warfare. Table and Figures | Reference | Related Articles | Metrics

Select

Energy efficiency optimization of the RIS-UAV communication system based on an improved TD3 algorithm WANG Yi, DENG Yu, XU Yaohua, JIANG Fang, JIANG Fulin, HU Yanjun Journal of Xidian University    2025, 52 (4): 226-234.   DOI: 10.19665/j.issn1001-2400.20250112 Abstract （254）   HTML （5）    PDF（pc） （1999KB）（41）       Knowledge map    Save Considering the presence of multiple mobile users in a Reconfigurable Intelligent Surface (RIS)-assisted Unmanned Aerial Vehicle (UAV) communication system,the impact of the UAV's flight energy consumption on the system's energy efficiency is investigated,with the system's energy efficiency improved by jointly optimizing the UAV trajectory and the active beam assignment as well as the RIS phase-shift design.The objective function is non-convex and the optimization variables are coupled,which makes it difficult for traditional algorithms to solve it directly.The present study proposes a solution in the form of a Gaussian Distribution Twin-Delayed Deep Deterministic Policy Gradient (GD-TD3) based on the Twin-Delayed Deep Deterministic Policy Gradient (TTD3).This is a method of optimising UAV trajectories in conjunction with UAV active beam assignments and RIS passive beam assignments.Its aim is to improve both the total system data rate and the long-term system energy efficiency.The proposed algorithm optimises UAV trajectories and the active/passive beam assignments of the UAV and RIS in the system by improving the original network structure in the two-intelligentsia framework while modeling the mobility of multiple users,respectively.Simulation results show that compared with other algorithms,the GD-TD3 algorithm performs better in the improvement of system energy efficiency,and that there is some improvement in the convergence speed and convergence stability. Table and Figures | Reference | Related Articles | Metrics

Select

PLM-RS:probabilistic logical reasoning for enhancing multi-modal sequential information in a recommender system WANG Jianfang, ZHU Xiangang Journal of Xidian University    2025, 52 (4): 120-133.   DOI: 10.19665/j.issn1001-2400.20250401 Abstract （246）   HTML （13）    PDF（pc） （2531KB）（44）       Knowledge map    Save Multimodal recommendation systems integrate multimodal information (e.g.,text,visual) into traditional recommendation frameworks to enhance the representation of users and items.This integration captures user interests and needs more accurately,thus providing more precise recommendations.However,recommendation tasks are not merely simple data induction problems; and they also require capturing users’ dynamic preferences and performing reasoning-based decision-making.To address this challenge,we propose a recommendation model (PLM-RS) that integrates multimodal sequential with probabilistic logical reasoning.To the best of our knowledge,we are the first to combine multimodal information with logical reasoning in a recommendation model.Specifically,we design an attention mechanism-based aggregator during the multimodal information fusion process.This aggregator computes attention scores to weight and average the embedding vectors in the input multimodal sequence,thereby enabling efficient aggregation of multimodal interactions to generate multimodal feature representations.Furthermore,we introduce an adjustable fusion module to dynamically balance the integration of multimodal features with sequential features captured by temporal modeling,so as to adapt to diverse recommendation scenarios.Additionally,we construct a logical loss function using alternating maximization and minimization of KL divergence to further enhance the model’s reasoning capabilities.Finally,we concatenate sequential multimodal features with probabilistic logical features to achieve the task of multimodal reasoning-based recommendations.Extensive experiments conducted on two public datasets demonstrate that the proposed method significantly improves multiple evaluation metrics for recommendation accuracy. Table and Figures | Reference | Related Articles | Metrics

Select

Profit-driven task scheduling algorithm for edge computing ZHANG Wuxuan, ZHAO Hui, ZHANG Guobin, WANG Jing, WAN Bo, WANG Quan Journal of Xidian University    2025, 52 (4): 77-92.   DOI: 10.19665/j.issn1001-2400.20250402 Abstract （235）   HTML （6）    PDF（pc） （1988KB）（34）       Knowledge map    Save Mobile Edge Computing (MEC) significantly enhances users’ computing and storage capabilities,yet its high construction and operational costs present a major challenge for service providers striving to maximize profit while maintaining the quality of service.Most existing studies focus solely on meeting task deadlines,neglecting the fact that users exhibit varying willingness-to-pay for different task completion times.This oversight complicates accurate revenue estimation and limits profit optimization.To address this gap,we propose a Profit Maximization Online Task Scheduling (PMOTS) algorithm for MEC environments.In our approach,we introduce the concept of expected task completion time and develop a dynamic task revenue model that captures a more nuanced relationship between completion time and service revenue.The PMOTS algorithm comprises two key components.First,the Serial Partitioned Computation Resource Allocation (SSCRA) algorithm recursively optimizes the execution order of task partitions upon each task’s arrival to maximize the incremental profit derived from resource allocation.Second,the Dynamic Allocation of the Idle Computation Resources (DAICR) algorithm prioritizes the assignment of idle computing resources to tasks with the highest expected profit increment,thereby reducing resource idleness and further enhancing overall profit.Simulation results demonstrate that the proposed approach effectively increases service provider profit in MEC environments. Table and Figures | Reference | Related Articles | Metrics

Select

Distributed dual access control scheme in cloud-edge-end environment ZHAO Yifan, ZHANG Jiawei, YANG Yanbo, HAN Lei, LI Teng, MA Jianfeng Journal of Xidian University    2025, 52 (4): 151-164.   DOI: 10.19665/j.issn1001-2400.20250201 Abstract （225）   HTML （2）    PDF（pc） （2027KB）（48）       Knowledge map    Save The development of emerging technologies such as the Internet of Things makes explosively growing various data a burden on resource-restrained end devices.Cloud Computing,centralized outsourcing and sharing services bring a long communication delay on time sensitive applications such as the Internet of Vehicles.Edge Computing offers low-latency service and promotes the development of the cloud-edge framework.However,storing and sharing data in the cloud-edge environment face fierce security threats,and effective dual access control provides basic protection for data safety.Although traditional attribute-based encryption could achieve grained dual access control,most of them neither offer access control over the data owner nor guarantee the reality and dynamic integrity of them.And these schemes also face the problems of data being tampered,node failure and key escrow.We propose a distributed and dynamic data integrity verifiable dual access control scheme with key escrow-free (KEF-DVDA),which uses KP-ABE and matchmaking encryption to implement fine grained access control over data users and data owners respectively while achieving the dynamic data integrity verification function by constructing a hierarchical Merkle Tree.Moreover,we also construct a key escrow-free multi-authorities key generation center which can distributes keys efficiently.A large number of simulation experiments have demonstrated the efficiency and practicality of the proposed scheme in this paper. Table and Figures | Reference | Related Articles | Metrics

Select

Aspect-based sentiment analysis of high-order dependencies and knowledge updating FAN Yating, HAN Hu, LI Lin, XU Xuefeng Journal of Xidian University    2025, 52 (4): 93-105.   DOI: 10.19665/j.issn1001-2400.20250404 Abstract （215）   HTML （9）    PDF（pc） （2279KB）（33）       Knowledge map    Save To address the challenges that it is difficult for external knowledge to be iteratively updated in specific domains and that existing methods often neglect variations in syntactic dependency distances within comments,this paper proposes an aspect-level sentiment analysis model based on higher-order dependencies and knowledge updating.First,two globally shared and dynamically updatable knowledge nodes are introduced to aggregate sentiment features from opinion words in comments,and then path-span information between word nodes is utilized to enhance the syntactic adjacency matrix integrated with external knowledge,thus enabling the model to capture multi-level,fine-grained dependency relationships between contexts and aspect terms and thereby dynamically optimizing sentiment weights for long-distance lexical nodes.Second,a multi-semantic and multi-dimensional distance module is constructed to learn local and global correlation features so as to effectively extract aspect-relevant semantic features,thereby achieving a deeper understanding of sentiment information related to aspect terms.Finally,the model adaptively fuses the learned sentence representations for sentiment prediction.Experimental results on five public datasets (Twitter,Lap14,Rest14,Rest15,and Rest16) show that the proposed method significantly outperforms baseline methods in accuracy and macro-F1 scores.Ablation and comparative experiments further validate the effectiveness of the higher-order dependency module in capturing sentiment-related information,and confirm that the knowledge updating strategy more precisely identifies fine-grained sentiment features. Table and Figures | Reference | Related Articles | Metrics

Select

Fusion time scale algorithm based on the Kalman filter LIU Qiang, SUN Haoran, ZHOU Jin, HU Denghua, ZHANG Shuang Journal of Xidian University    2025, 52 (4): 248-260.   DOI: 10.19665/j.issn1001-2400.20250106 Abstract （215）   HTML （5）    PDF（pc） （2752KB）（34）       Knowledge map    Save As an important component of the timekeeping system,the time scale algorithm’s core task is to establish and maintain a stable,accurate,and reliable time scale for the timekeeping system.In response to the problem that the traditional weighted average time scale algorithm lacks attention to noise when calculating the time scale and the real-time requirements of the timekeeping system,this paper proposes a time scale algorithm based on the Kalman filter and the fusion of time deviation and frequency deviation residuals to calculate the weights.After the Kalman filter algorithm calculates the optimal estimates of the time deviation and frequency deviation of the atomic clock,the algorithm fuses the time deviation residuals and frequency deviation residuals to calculate the weights of the atomic clock.Finally,the time scale is obtained by weighted averaging using the basic equation of the time scale.The advantage of the time scale algorithm that calculates the weights by the fusion method is that it comprehensively considers the influence of time deviation and frequency deviation on the time scale,and improves the real-time performance of the algorithm with the help of the Kalman filter.Through simulation verification,it is found that the short-term stability of the time scale calculated by the time scale algorithm that calculates the weights by the fusion method is better than that of the time scale calculated by the time scale algorithm that only calculates the weights of the atomic clock based on time deviation,ALGOS,AT1 algorithm,and the traditional Kalman filter time scale algorithm,and that the proposed algorithm has a real-time performance. Table and Figures | Reference | Related Articles | Metrics

Select

Ground-air handover authentication scheme for ATN based on the Chinese remainder theorem SHANG Tao, TIAN Gege, JIANG Yatong, LIU Zhentao Journal of Xidian University    2025, 52 (4): 208-225.   DOI: 10.19665/j.issn1001-2400.20241106 Abstract （207）   HTML （5）    PDF（pc） （2650KB）（19）       Knowledge map    Save With the rapid development of civil aviation,the demand for air-to-ground data communication is increasing.As the global ground-air integrated dedicated network,the aeronautical telecommunications network integrates data communication networks and services with different needs,including air traffic control service and airline operation center in civil aviation.At the beginning of design,the aeronautical telecommunications network lacks the corresponding security mechanism and faces significant security challenges in the complex,heterogeneous ground-to-air communication environment.To meet the high stability requirements of the heterogeneous aeronautical telecommunication network handover authentication,a handover authentication protocol based on the Chinese Remainder Theorem is proposed for the aeronautical telecommunication network.The Chinese Remainder Theorem is utilized to conduct key agreements for multiple potential access nodes at a time,enhancing stability during handover authentication.Based on the architecture of the cockpit network COMmunications Environment Testing,the Software-Defined Networking controller is expanded to solve problems such as the lack of security mechanisms,cross-domain handover,and multi-link access.Multiple signaling processes are merged to simplify the handover authentication steps,reducing communication overhead,with this protocol lowering the computational overhead by using Chaotic maps.The security properties of the protocol are analyzed by the formal analytical tool AVISPA,and its performance is compared with other solutions,which proves that it can ensure stable and efficient handover authentication. Table and Figures | Reference | Related Articles | Metrics

Select

New explanation of digital implementation of keystone transformation SHI Hao, LIAO Guisheng Journal of Xidian University    2025, 52 (4): 235-247.   DOI: 10.19665/j.issn1001-2400.20250303 Abstract （204）   HTML （8）    PDF（pc） （3316KB）（24）       Knowledge map    Save Despite the widespread application of the Keystone transform in radar signal processing,several issues persist in its digital implementation.First,from the perspective of interpolation theory,the selection of new sampling points can be ambiguous,and the frequency-domain aliasing caused by downsampling is often overlooked.Second,the relationships among existing digital implementation methods of the Keystone transform have not been thoroughly analyzed.To address these issues,this study reformulates the digital implementation of the two-dimensional Keystone transform as a one-dimensional sequence resampling problem.Based on the interpolation theory and integrating the fractional resampling theory,we derive multiple sequence resampling implementation methods from a spectral analysis perspective.The interconnections among these methods are analyzed,thereby enhancing the theoretical framework for the digital implementation of the Keystone transform and providing a reference for a deeper understanding of its principles and methodologies.Furthermore,leveraging the principles of the Chirp-Z transform fast algorithm,we analyze the computational complexity of two fast algorithms,offering a clear and standardized metric for the computational load of the Keystone transform.Finally,simulation results validate the correctness of the analysis and the effectiveness of the proposed algorithms. Table and Figures | Reference | Related Articles | Metrics

Select

Post-quantum secure privacy enhancement scheme for gas industrial IoT FAN Jianyong, MA Guanyong, TAN Maolin, MU Dejun, TANG Bo, LIU Jinhui Journal of Xidian University    2025, 52 (4): 192-207.   DOI: 10.19665/j.issn1001-2400.20250708 Abstract （192）   HTML （4）    PDF（pc） （2218KB）（25）       Knowledge map    Save In order to deal with the security risks of traditional public key cryptographic algorithms under the threat of quantum computing and the difficulties in behavioral auditing in the adaptation process of gas IoT terminals such as NB-IoT smart gas meters,this paper proposes a post-quantum security and privacy protection enhancement scheme for the gas industrial IoT.This scheme introduces the SPHINCS+ signature algorithm and the Boojum-based zero-knowledge concise non-interactive knowledge argumentation mechanism to achieve efficient identity authentication and data integrity verification without trusted settings.At the same time,combined with the group signature mechanism,on the basis of ensuring user anonymity,it supports the identity revocability function in regulatory scenarios and improves the audit and accountability capabilities of the system.The experiment is carried out in a resource-constrained NB-IoT device and a high-concurrency edge computing environment,focusing on evaluating key performance indicators such as communication load,signature overhead and verification efficiency.Simulation results show that while achieving post-quantum security,the scheme significantly reduces communication and computing overheads,and has stronger adaptability and scalability than traditional schemes.The study shows that the privacy protection enhancement scheme can effectively meet the comprehensive requirements of the gas industrial IoT for security,privacy and regulatory auditability,and provides a practical technical path for building a new gas IoT system that is trustworthy,controllable and available. Table and Figures | Reference | Related Articles | Metrics

Select

Study of passive location based on the fertile field algorithm HOU Xiaogeng, DU Qinghe, MIAO Zhongyu, CHEN Xiaoyang, XIAO Yuquan, LIU Feng Journal of Xidian University    2025, 52 (4): 46-54.   DOI: 10.19665/j.issn1001-2400.20250705 Abstract （184）   HTML （7）    PDF（pc） （1349KB）（44）       Knowledge map    Save In this article,the high-precision and high-stability passive location based on fertile field algorithm (FFA) is studied.First,multi-station passive location nonlinear equations for ground targets are derived,and then they are transformed to an unconstrained objective function in a penalty-based way,making it easier for them to be solved through optimization methods than by direct equation-solving approaches.For minimizing the objective function to obtain the globally optimal solution,a multi-station passive location method based on the fertile field optimization algorithm is proposed which integrates global exploration and local exploitation by leveraging mechanisms such as wind and animals to facilitate seed dispersal and high-fertility seed selection.While avoiding premature convergence to local optima,the approach progressively converges to the global optimal solution,demonstrating robust global convergence properties.Since the solution for target position does not rely on any initial conditions,the proposed method enables precise location estimation in passive location systems without requiring coarse initial value estimation,thus demonstrating high universality.It is also verified that the proposed algorithm can achieve global convergence with fewer iterations,especially having a relatively high stability during multiple positioning of different targets.In addition,the high location accuracy can be obtained under different requirements on the estimation error.Based on these advantages,the proposal can be used to analyze and judge the situation of massive targets. Table and Figures | Reference | Related Articles | Metrics

Select

Brain-controlled wheelchair integrating the Savitzky-Golay filter and an extended canonical correlation analysis PAN Hongguang, TENG Bingyang, YU Xinyu, ZHANG Tuo, MI Wenyu Journal of Xidian University    2025, 52 (4): 134-150.   DOI: 10.19665/j.issn1001-2400.20250113 Abstract （178）   HTML （8）    PDF（pc） （5412KB）（28）       Knowledge map    Save The Brain-Controlled Wheelchair (BCW) integrates Brain-Computer Interface (BCI) technology with an electric wheelchair,empowering individuals with motor impairments to maneuver the wheelchair by using their thoughts,thereby enhancing their quality of life.However,the efficacy and practical applicability of the existing BCW system necessitate urgent enhancement.This paper integrates steady-state visual evoked potentials and electrooculography (EOG) to develop an embedded asynchronous BCW system,aiming to enhance its overall performance and practical utility.First,by leveraging the distinct characteristics of EOG waveform peaks and troughs,the slope threshold method is employed for real-time detection of blink events,thereby provisioning an asynchronous control mechanism for the initiation and cessation of the BCW system.Second,a Savitzky-Golay filter based wavelet packet decomposition is proposed for the smoothing and filtering of EEG signals,with its parameters optimized through grid search,effectively eliminating various low-frequency motion artifacts while preserving the original information.Employing extended canonical correlation analysis to identify the frequency components that represent specific visually evoked activities within the signal,and constructing signal templates and artificially generated reference signals by using offline datasets to comprehensively compute signal correlations and decode them,thereby providing accurate control commands for the BCW system.Finally,the proposed algorithm is integrated into embedded devices,with the effectiveness of the embedded BCW system verified through experimental validation.Experimental results show that through the online BCW evaluation experimental results the average accuracy of blink event detection in the straight line test scene is 83.29%,that the average classification accuracy of online EEG signals is 82.93%,and that the task completion rate is up to 87.5%.The average accuracy of blink event detection in complex environment test scenarios is 83.66%,the average classification accuracy of online EEG signals is 81.75%,and the task completion rate is up to 62.50%.This study enhances the overall performance and practicality of BCW,laying a crucial foundation for its commercialization and everyday use. Table and Figures | Reference | Related Articles | Metrics

Select

Novel anti-TARD method for terminal guidance radar YE Qingzhi, CHEN Baixiao, ZHANG Yan, ZHU Shengqi, DENG Zhao Journal of Xidian University    2025, 52 (5): 13-25.   DOI: 10.19665/j.issn1001-2400.20250704 Abstract （177）   HTML （31）    PDF（pc） （2735KB）（91）       Knowledge map    Save As an active deceptive jamming style, the towed Radar Active Decoy(TRAD) plays a crucial role in modern electromagnetic spectrum warfare. These TRADs modulate and retransmit radar signals to simulate target echoes, inducing radars to track stronger false targets, thereby providing self-defense protection for the aircraft. To address the vulnerability of terminal guidance radars to TRAD deception, this paper proposes a countermeasure method based on spatial morphological features. Since jamming signals cannot realistically replicate the spatial morphology of targets under wideband conditions, the proposed method enables effective target discrimination and anti-jamming imaging. First, a monopulse three-dimensional(3-D) imaging algorithm is employed to reconstruct the spatial distribution of strong scatterers in the radar's forward-looking region. Second, spatial filtering is applied to the imaging results to remove noise points and cluster the strong scattering point clouds. Finally,spatial morphological features are extracted from each point cloud cluster. A discriminator is then employed to perform target identification, resulting in an anti-jamming three-dimensional image of the target. Theoretical analysis and experimental results reveal significant spatial morphological differences between the false targets produced by the TRAD and the actual targets. These findings validate the effectiveness of the proposed method in countering TRAD jamming. Table and Figures | Reference | Related Articles | Metrics

Select

Method for recognizing multifunctional radar working modes based on the ST-GCN SUN Yuwen, XIE Rong, WANG Yefei, XU Shuwen, LIU Zheng Journal of Xidian University    2025, 52 (5): 121-132.   DOI: 10.19665/j.issn1001-2400.20250807 Abstract （125）   HTML （8）    PDF（pc） （2780KB）（24）       Knowledge map    Save Multifunction radar (MFR) realizes multi-task coordination through waveform agility and beam adaptive scheduling, which brings many challenges to radar working pattern recognition. Existing recognition methods rely on the local time-domain characteristics of the pulse sequence, and it is difficult to effectively analyze the generation mechanism of different working modes. In the face of complex situations such as pulse loss and similar intra-pulse parameters, the recognition performance drops sharply. Considering the influence of multi-function radar beam scanning process on the amplitude information of pulse group sequence, a multi-function radar working pattern recognition method based on the Spatial-Temporal Graph Convolutional Network (ST-GCN) is proposed. The network model first quantifies the similarity of radiation characteristics of adjacent wave position signals by introducing a dynamic regularization module, and constructs a spatial adjacency matrix with physical interpretability. Then, the one-dimensional pulse group sequence is mapped into a two-dimensional graph structure, and the node characteristics such as pulse frequency and signal amplitude are fused to form a space-time joint representation. Finally, the convolutional kernel of the layered graph is designed with the deep space-time features extracted through the multi-layer information transmission mechanism to complete the radar working pattern recognition. Comparative experiments show that the average recognition rate of the proposed method can still reach 93.38 % under non-ideal conditions such as pulse loss, and can be better generalized and robust. Table and Figures | Reference | Related Articles | Metrics

Select

Semantic communication system for text-based screen content image transmission ZHU Zhiyuan, LI Jie, TANG Tong, LV Yi, WU Dapeng, WANG Ruyan Journal of Xidian University    2025, 52 (5): 26-38.   DOI: 10.19665/j.issn1001-2400.20250701 Abstract （117）   HTML （22）    PDF（pc） （1922KB）（56）       Knowledge map    Save The text-based screen content is widely used in scenarios such as online meetings, cloud gaming, and remote control, and it generates massive data that impose a significant pressure on storage and bandwidth. Traditional screen content transmission methods are constrained by the Shannon limit and cannot break through the transmission bottleneck. Moreover, existing image coding and transmission methods based on semantic communication mainly target natural scene images, with screen content images not sufficiently studied. In response to the above issues, a semantic coding and transmission system for text-based screen content images is constructed for the first time. Specifically, an end-to-end semantic framework is first designed to extract the semantic features of textual information in images while compressing or discarding irrelevant data. By reducing the amount of unnecessary data transmitted, efficient encoding of key semantic features is achieved. Next, a dynamic encoding strategy utilizing the channel state information (CSI) feedback is proposed, which optimizes the encoding process by real-time acquisition of wireless channel characteristics to enhance transmission robustness and efficiency. Additionally, an auxiliary loss function is designed to protect semantic features relevant to downstream tasks in the images. Finally, experimental results demonstrate that compared with the state-of-the-art transmission methods, the proposed approach achieves a higher communication efficiency under different channel conditions and bit rates. Table and Figures | Reference | Related Articles | Metrics

Select

CITree:a hybrid packet classification algorithm based interval tree LI Zhuo, PEI Yuheng, XUN Hao, LIU Jindian Journal of Xidian University    2025, 52 (5): 48-58.   DOI: 10.19665/j.issn1001-2400.20250505 Abstract （103）   HTML （13）    PDF（pc） （2116KB）（44）       Knowledge map    Save High-performance classification throughput and dynamic update of rule sets are two core requirements for packet classification algorithms. Hybrid packet classification algorithms combining hash tables and decision trees typically replace tree nodes with hash tables to compensate for the inherent shortcomings of decision trees in terms of update performance. However, the hash table instead of tree nodes introduces additional hash mappings in the packet classification and disrupts the original single query path of the tree structure, ultimately degrading the classification performance. Consequently, a hybrid packet classification algorithm based on the interval tree, named CutIntervalTree (CITree), is proposed to achieve a high classification performance while supporting fast rule updates. The algorithm avoids introducing hash tables into the tree structure and instead introduces a preprocessing unit to partition the rule set into independent and non-independent rules, and stores them in the two-level tree structure and the hash table, respectively. The categorized storage of rules fully utilizes the advantages of high-speed classification of tree structure and fast update of the hash table. In addition, the CITree stores independent rules in the root node, intermediate nodes and leaf nodes of the interval tree, so that the matching operation can be completed in any node without traversing to the leaf nodes, thus realizing effective pruning. Experimental results show that the proposed algorithm improves the classification throughput by 72.2% and the rule updating efficiency by 63% compared with the current state-of-the-art algorithm. Table and Figures | Reference | Related Articles | Metrics

Select

Density peak clustering combining cluster growth and boundary assignment strategy CHEN Sugen, ZHAO Zhizhong Journal of Xidian University    2025, 52 (5): 143-161.   DOI: 10.19665/j.issn1001-2400.20250801 Abstract （99）   HTML （5）    PDF（pc） （3197KB）（21）       Knowledge map    Save Aiming at the problems that the density peak clustering algorithm has a poor clustering effect on variable density datasets and that the "domino" phenomenon will occur in the sample assignment process, a density peak clustering algorithm combining cluster growth and boundary assignment strategy is proposed. The algorithm uses the local k-nearest neighbor information to calculate the sample density and relative distance, and then obtains the sample decision value. Based on the distance, density and neighbor relationship between samples, the attraction degree and growth radius are defined. Combined with the decision value, the cluster centers are selected in turn, and the cluster growth strategy is proposed. Starting from each cluster center, this strategy grows the current cluster by using the attraction degree and the growth radius to obtain the initial clustering result, on the basis of which the adjacency degree is defined by using the nearest neighbor and distance information between the assigned clusters and the unassigned samples, and the boundary assignment strategy is proposed. The assignment strategy divides each unassigned sample into the most appropriate cluster by the adjacency degree, and updates the assigned and unassigned sample sets continuously until all the samples are assigned to obtain the final clustering result. Compared with 7 algorithms on 16 synthetic datasets and 10 UCI datasets, experimental results show that the proposed algorithm is superior to the comparison algorithms in adjusted rand index, normalized mutual information and adjusted mutual information on most datasets. At the same time, the statistical test results show that the proposed algorithm and the comparison algorithm have statistically significant differences. The proposed algorithm has a better clustering effect. Table and Figures | Reference | Related Articles | Metrics

Select

Text-guided multi-modal mural diffusion restoration method CHEN Yong, ZHANG Shilong Journal of Xidian University    2025, 52 (5): 133-142.   DOI: 10.19665/j.issn1001-2400.20250804 Abstract （97）   HTML （9）    PDF（pc） （6411KB）（25）       Knowledge map    Save The objective of digital mural restoration is to employ information technology in reconstructing the missing or damaged portions of murals, thereby reinstating their visual coherence and authentic artistic representation. Existing deep learning methods for mural restoration often lack sufficient cross-modal semantic constraints from the text, which can lead to semantic confusion and loss of detail in the restoration results. To address this issue, we propose a text-guided multi-modal mural diffusion restoration method. First, a text encoding module based on a multi-head self-attention mechanism is designed to project mural textual descriptions into the feature space. A cross-modal interaction mechanism is further introduced to fuse textual and visual features, thereby enhancing the semantic consistency between modalities. Then, we build a diffusion-based mural restoration module. The forward diffusion process adds noise to generate Gaussian-distributed mural features, while the reverse network reconstructs missing regions. Next, a mask refinement control module is introduced. It uses features from the complementary mural mask to guide the reverse decoding process and improve the generation of texture and details, enabling accurate restoration of damaged murals. Finally, experiments on the Dunhuang mural dataset show that the proposed method outperforms comparison methods. Table and Figures | Reference | Related Articles | Metrics

Select

Pilot-aided symbol synchronization and channel estimation techniques for AFDM systems GUO Lin, GUO Wei, WANG Yongchao Journal of Xidian University    2025, 52 (5): 59-71.   DOI: 10.19665/j.issn1001-2400.20250602 Abstract （93）   HTML （8）    PDF（pc） （1506KB）（34）       Knowledge map    Save As a multicarrier modulation technique proposed for high-speed mobile scenarios, AFDM can completely separate each path of the doubly selective channel, achieving full time and frequency diversity gains. It is a strong candidate waveform for the physical layer of future mobile communications. However, current research achievements on symbol synchronization for AFDM systems are very limited, and traditional synchronization algorithms struggle to achieve a good performance in complex doubly selective channels. Aiming at this problem, this paper proposes a coarse-fine two-stage synchronization algorithm suitable for doubly selective channels. By relying on fine synchronization performed in the discrete affine Fourier transform(DAFT) domain to correct errors from coarse synchronization, a high synchronization accuracy is obtained under doubly selective channels, on the basis of which this paper combines the repeated multi-segment pilot structure required for coarse synchronization, and jointly designs fine synchronization with channel delay-Doppler shift estimation, thus saving pilot overhead. Finally, the channel complex gain information is obtained using the LS estimation algorithm, thereby realizing symbol synchronization and channel estimation for the AFDM system. Simulation results show that the algorithm has a high synchronization accuracy and an accurate channel estimation in time-frequency dual selection channels, and that the requirement for pilot energy in channel estimation is significantly reduced, demonstrating its good adaptability and application prospects in time-frequency dual selection channel environments. Table and Figures | Reference | Related Articles | Metrics

Select

Research on indoor 3D VLP using the multi-strategy improved Jaya algorithm HE Huimeng, WANG Ping, LIU Qi, YANG Ting, SUN Yanzhe Journal of Xidian University    2025, 52 (5): 39-47.   DOI: 10.19665/j.issn1001-2400.20250511 Abstract （93）   HTML （14）    PDF（pc） （1645KB）（30）       Knowledge map    Save An indoor visible light positioning (VLP) scheme based on the multi-strategy improved Jaya algorithm is proposed for the indoor 3D VLP problem of a single light-emitting diode (LED), taking into account the influence of non-line-of-sight (NLOS) links. Specifically, in this indoor 3D VLP system model, the NLOS link is considered as an interference, and the PAM-DMT-based least squares (LS) channel estimation is utilized to estimate the LOS link channel gain, and then the sum of the squares of the difference between the estimated channel gain and the channel gain computed at the test point is used as the fitness function, and the multi-strategy improved Jaya algorithm is used for the search and optimization to realize the localization. Subsequently, the 3D positioning errors of different numbers of PDs at different inclination angles and heights are simulated and compared, and the convergence and performance of the proposed multi-strategy improved Jaya algorithm are compared with those of the other four algorithms. The results show that the LS channel estimation method based on PAM-DMT can effectively reduce the localization error due to NLOS interference, and that the average localization error of the system with channel estimation is reduced by 84.64% compared with that of the system without channel estimation. When the number of PDs in the receiver is 5 and the inclination angle is 75°, the fluctuation of the positioning error with height is the lowest and the average positioning error is the smallest under the consideration of the NLOS link. Compared with the other four algorithms, the multi-strategy improved Jaya algorithm has a higher positioning accuracy as well as fewer iterations. The work in this paper will be of some reference value for the study of the indoor single-LED VLP system. Table and Figures | Reference | Related Articles | Metrics

Select

Analysis of privacy leakage of the mini program based on covert communication LIU Lipei, MAO Jian, LIN Qixiao, LV Yusong, LI Jiawei, LIU Jianwei Journal of Xidian University    2025, 52 (5): 173-182.   DOI: 10.19665/j.issn1001-2400.20250706 Abstract （91）   HTML （8）    PDF（pc） （1770KB）（12）       Knowledge map    Save Mini programs, exemplifying the "app-in-app" paradigm, have become deeply integrated into people's work and daily lives, accessing substantial amounts of user privacy data. To prevent privacy leaks, mini program platforms monitor and regulate regular communication methods. However, mini programs can use covert communication to evade detection. Aiming at the security threat of covert communication to user privacy leakage, this paper analyzes the risk of privacy leakage of mini programs covert communication. On the basis of summarizing the covert communication model and communication conditions of mini programs, we design covert communication methods for both mini-program-to-mini-program and mini-program-to-server communications based on the mini program APIs and components. Invisible character-based source coding and forged pages are adopted to improve the covertness respectively. Experiments verify that the above covert communication methods can realize secret information transmission, and that two attack scenarios are designed to analyze the privacy leakage risk brought by the covert communication methods. Finally, corresponding mitigation measures are discussed. Table and Figures | Reference | Related Articles | Metrics

Select

New method for wide-range frequency measurement using digital linear phase comparison LI Zhiqi, GUO Yi, CHEN Xiaolong, ZHU Jiahao Journal of Xidian University    2025, 52 (5): 111-120.   DOI: 10.19665/j.issn1001-2400.20250901 Abstract （89）   HTML （12）    PDF（pc） （1845KB）（20）       Knowledge map    Save Aiming at the current problems that although the direct counting frequency measurement has a wide measurement range, its resolution is limited, and although the indirect phase measurement has a high resolution, it has measurement dead zones and a narrow measurement range. This paper proposes a digital linear phase comparison method with a variable least common multiple period for frequency and frequency stability measurement. This method employs an analog to digital converter as a phase detector to acquire phase information, and the digital measurement approach avoids the measurement dead zone existing in traditional phase comparison methods. By controlling the phase detection region to be within the linear range, the measurement ambiguity zone is circumvented, and the measurement resolution is improved. By analyzing the variation law of phase difference between signals with arbitrary frequencies, taking the nominal least common multiple period as the sampling interval, and combining rough frequency measurement with truncation processing, the measurement range is expanded, and the noise introduced during the normalization process is avoided. Data processing is performed on the phase information in the linear region obtained by sampling, thereby completing the direct phase comparison between signals of arbitrary frequencies and achieving high-resolution measurement of frequency and frequency stability. Experimental results show that the system noise floor is better than 1.86E-15/1000s, and that under the measurement gate of 1 second, the resolution of frequency measurement can reach 20μHz. When performing the high-resolution frequency and frequency stability measurement on frequency source signals in the range of 1MHz - 100MHz, the accuracy of the measurement results remains stable. Table and Figures | Reference | Related Articles | Metrics

Select

Circuitdesign and effective simulation of a class of quantum solving algorithm GAO Jiapei, LI Xuelian, GAO Juntao Journal of Xidian University    2025, 52 (5): 72-87.   DOI: 10.19665/j.issn1001-2400.20250504 Abstract （83）   HTML （12）    PDF（pc） （2204KB）（29）       Knowledge map    Save Classical simulation of quantum algorithms plays a crucial role in evaluating the algorithm performance and verifying theoretical correctness. For high-order sparse matrices, the corresponding Hamiltonians often exhibit complex structures and characteristics, leading to excessively high complexity in quantum solving, which severely constrains simulation efficiency and accuracy. To address the challenges in simulating Hamiltonians, modular decomposition techniques and function construction methods are proposed to approximate the evolution of Hamiltonians, thereby establishing a general circuit design scheme for implementing the Harrow-Hassidim-Lloyd(HHL) algorithm on classical computers. We implement multi-scale quantum circuits with 13/14 qubits (basic scale) and 20/21 qubits (extended scale) based on the Qiskit quantum computing framework, and verify the applicability of the designed circuits by testing multiple sets of 8×8 Hermitian matrices and column vectors. Finally, we analyze the fidelity and error under different conditions for various linear systems, as well as the time and space resources they occupy. Experimental results demonstrate that as the qubit scale expands, quantum circuits incorporating these two techniques exhibit synchronous optimization characteristics with enhanced fidelity and reduced errors when solving linear systems. Compared with other methods, both techniques demonstrate superior large-scale circuit processing capabilities, providing a scalable technical route for utilizing quantum algorithms to solve high-dimensional linear systems. Table and Figures | Reference | Related Articles | Metrics

Select

Collaborative attention mechanism-based iris segmentation algorithm LI Yongbo, DU Jianchao, WANG Junning, ZHANG Mingjin Journal of Xidian University    2025, 52 (5): 88-98.   DOI: 10.19665/j.issn1001-2400.20250806 Abstract （83）   HTML （12）    PDF（pc） （3362KB）（29）       Knowledge map    Save Aiming at the problem of poor performance of iris segmentation for low-quality images, this paper proposes an iris segmentation algorithm based on the collaborative attention mechanism. Based on the U-Net model under the deep learning framework, this algorithm innovatively introduces a dual attention module of the regional attention mechanism and the quality-aware attention mechanism, and collaboratively improves the accuracy of iris region segmentation through the two dimensions of position perception and image quality perception. Specifically, the regional attention mechanism predicts the area where the iris ring is located and constrains the target spatial area during the feature extraction process, thereby effectively reducing background noise interference. The quality-aware attention mechanism dynamically adjusts the focus on key features in the convolutional attention module based on the image quality assessment results, thereby significantly enhancing the key feature expression ability of low-quality images. Experimental results on public datasets and self-made datasets show that this algorithm outperforms many mainstream segmentation models such as U-Net and IrisParseNet in the two core segmentation evaluation metrics of intersection and union ratio and accuracy rate. Especially under the conditions of low-quality images such as low illumination and motion blur, the improvement of segmentation effect is more significant. These improvements provide reliable technical support for the practical application of iris recognition systems in complex environments. Table and Figures | Reference | Related Articles | Metrics

Select

Enhancing social recommendation via the knowledge-aware neighbor filtering mechanism ZHOU Jialiang, MU Caihong, LIU Yi, CHEN Yunlong Journal of Xidian University    2025, 52 (5): 99-110.   DOI: 10.19665/j.issn1001-2400.20250603 Abstract （81）   HTML （5）    PDF（pc） （3577KB）（23）       Knowledge map    Save Social recommendation helps to improve the performance of personalized recommender systems by exploiting user social connections. However, most existing methods struggle to fully capture the complex relationships between users and items, while neglecting the issue of social inconsistency caused by irrelevant or even erroneous social ties, thus reducing the correctness of user embeddings and the accuracy of social recommendations. This paper proposes a knowledge-aware neighbor filtering mechanism for social recommendation (KFRec), aiming to resolve the aforementioned issues by integrating knowledge graphs with graph neural networks. First, this paper utilizes knowledge graph embedding techniques to vectorially represent users, items, and ratings, thereby capturing the latent relational patterns among them. Subsequently, these vectors are fed into a graph neural network to optimize the node representations of the graph neural network. To improve the model's consistency recognition capability, this paper dynamically constructs query vectors based on the user-item pairs to be evaluated, and comprehensively model the consistency scores between the query vectors and neighbor nodes using the knowledge graph. By sampling and aggregating more consistent neighbor nodes, the graph neural network model's ability to filter inconsistent neighbor nodes and node representations is enhanced. Extensive experiments on three public datasets demonstrate the superiority of our proposed KFRec over existing mainstream methods. Table and Figures | Reference | Related Articles | Metrics

Select

Research on the attack algorithm for prime-field ECDLP using D-Wave quantum computers WANG Chao, YANG Shuxiao, PEI Zhi, HONG Chunlei, LI Yu, HAN Yiliang, ZHU Shuaishuai Journal of Xidian University    2025, 52 (5): 205-216.   DOI: 10.19665/j.issn1001-2400.20250501 Abstract （81）   HTML （10）    PDF（pc） （1480KB）（29）       Knowledge map    Save Elliptic Curve Cryptography is a class of public-key cryptographic algorithms with exponential attack difficulty, widely used in fields such as the encryption of second-generation ID cards. Shor’s algorithm theoretically poses a fatal threat to public-key cryptography, but to date, there have been no reports in the open literature on successful applications of quantum algorithms to attack ECC. In response to the current gap in quantum algorithms for ECC attacks, this paper proposes a quantum annealing-based algorithm for attacking the Elliptic Curve Discrete Logarithm Problem over finite fields. The approach begins by optimizing the coefficients in the Ising model conversion process during quantum annealing, significantly reducing the weight and coupling strengths (hi and Ji,j) of the relevant qubits by over 89.02%. By using quantum annealing to solve the Ising model optimized with the Semaev summation polynomial, the energy gap during the annealing process is greatly reduced, thereby revealing the relationships between points on the elliptic curve. Next, a sufficient number of Semaev polynomials are solved, and the resulting relationships are transformed into a system of linear equations. For this transformed linear system, a new algorithm based on quantum annealing is proposed for solving linear equations, enabling the solution of underdetermined and non-square linear systems. Ultimately, this work successfully solves the ECDLP over a finite field of up to 10-bits using the D-Wave Advantage, achieving a finite field size that is 289% larger than that of the previous largest solution. Experimental results show that the proposed method can effectively reduce the solution difficulty of D-Wave quantum annealing, and is a new quantum algorithm that can effectively attack ECDLP. Table and Figures | Reference | Related Articles | Metrics

Select

Optimal weight function sparse array with a low coupling effect HAN Yujie, WANG Lanmei, CHE Yu, WANG Guibao Journal of Xidian University    2025, 52 (5): 162-172.   DOI: 10.19665/j.issn1001-2400.20250707 Abstract （73）   HTML （5）    PDF（pc） （2163KB）（22）       Knowledge map    Save In the field of array signal processing, existing improved arrays have significant performance limitations: although the improved coprime array has a sparse structure, its ability to fill holes is weak, resulting in a limited number of identifiable incoming signals. Although the improved nested array can fill most of the array holes to support the identification of more incoming signals, the densely distributed array elements cause strong coupling effects, which seriously restrict the angle estimation accuracy. To address the above contradictions, this paper proposes an Optimal Weight function Sparse Array (OWSA) based on a newly improved array configuration. By optimizing the array element layout and weight allocation, this array further improves the sparsity of array elements on the basis of ensuring a long uniform degree of freedom, thereby effectively alleviating the electromagnetic coupling effect between array elements and achieving the synergistic improvement of the ability to identify multiple incoming wave angles and the coupling suppression performance. Experimental results show that the OWSA array successfully balances the core problem that it is difficult for the uniform degree of freedom and coupling effect to be taken into account in traditional array configurations. In complex scenarios such as a low signal-to-noise ratio, a small number of snapshots, and high coupling, its direction-of-arrival angle estimation accuracy is significantly better than that of the improved coprime array and nested array, which verifies the feasibility and superiority of this new array in high-precision angle estimation in complex electromagnetic environments. Table and Figures | Reference | Related Articles | Metrics

Select

Steganalysis scheme for combining element-wise multiplication and orthogonal fusion WANG Changguang, SHI Haoyi, LI Qingru, WANG Fangwei Journal of Xidian University    2025, 52 (5): 183-192.   DOI: 10.19665/j.issn1001-2400.20250502 Abstract （68）   HTML （5）    PDF（pc） （1896KB）（9）       Knowledge map    Save One of the main challenges in image steganalysis is to maintain a high detection accuracy while simplifying the model structure, including reducing the number of trainable parameters and accelerating both training and inference. To address this issue, this paper proposes a steganalysis scheme that combines element-wise multiplication with deep orthogonal fusion. First, a multi-scale attention module is designed to enhance the noise residual features extracted by the SRM filters during the preprocessing stage. Then, a feature analysis module incorporating separable convolutions and element-wise multiplication is introduced to perform multi-scale modeling and learning of the enhanced noise features. Finally, an orthogonal feature fusion module is proposed to integrate local and global features of the noise in an orthogonal manner, compensating for the loss of fine-grained details caused by global average pooling in the analysis process. Experiments are conducted on two public datasets, BOSSBase and BOWS 2, using two typical adaptive steganographic algorithms, S-UNIWARD and WOW, at various embedding rates. The experimental results demonstrate that the proposed method achieves approximately 2.4% and 1.2% higher detection accuracies on average compared to commonly used methods for the two algorithms, while significantly reducing the number of model parameters and improving an overall efficiency and performance. In addition, ablation studies validate the effectiveness of each proposed module within the overall framework. Table and Figures | Reference | Related Articles | Metrics

Select

Blockchain-assisted SM2 secure communication protocol for VANET CAO Peiwei, SONG Kai, YANG Wenjie, ZHANG Futai, LI Qing Journal of Xidian University    2025, 52 (5): 193-204.   DOI: 10.19665/j.issn1001-2400.20250802 Abstract （58）   HTML （9）    PDF（pc） （1442KB）（19）       Knowledge map    Save The Vehicular Ad Hoc Network (VANET) is a specialized form of the mobile ad hoc network that utilizes vehicles and infrastruc-ture as nodes. It uses wireless communication technology to exchange data and share information among these nodes, thereby significantly optimizing traffic efficiency, ensuring driving safety, and improving user experience. However, trans-mitting data in a plaintext during communication exposes it to various security attacks such as data forgery and theft. Existing VANET communication mechanisms prioritize data authenticity over confidentiality. While these mechanisms ensure the au-thenticity of transmitted data, they may not provide sufficient protection for its confidentiality. Furthermore, these mechanisms encounter challenges in promptly identifying and revoking access for malicious vehicles that can compromise network security. To tackle these problems, this paper condenses the core technology of the SM2 algorithm, designs an efficient sign-cryption scheme based on SM2, and its security is proven under the random oracle model. The signcryption scheme is then implemented in the VANET system and integrated with blockchain smart contract technology to manage vehicle certificates. This VANET protocol guarantees both data confidentiality and authenticity in a single logical step, while also achieving rapid tracking and revocation of malicious vehicles. The introduction of smart contract further improves the transparency and credibility of the system. Experimental analysis demonstrates that, compared with the existing work, the proposed protocol reduces the communication overhead by about 12% without significantly increasing the computational overhead. Table and Figures | Reference | Related Articles | Metrics