检索结果-内蒙古大学图书馆

A Lightweight SAR Ship Detection Network Based on Deep Multiscale Grouped Convolution, Network Pruning, and Knowledge Distillation

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2025年 18卷 2190-2207页

作者： Hu, Boyi Miao, Hongxia Beijing Univ Posts & Telecommun Sch Informat & Commun Engn State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

Deep learning has proven to be highly effective in synthetic aperture radar (SAR) image target detection. However, many latest deep learning models have predominantly focused on increasing depth and size to enhance detection accuracy, often ignoring the balance between accuracy and detection speed, as well as the practical deployment of these models on hardware platforms. Therefore, a lightweight algorithm for SAR ship detection is designed in this article. First, we propose a preliminary lightweight scheme, including a multiscale feature learning augmented backbone, a lightweight feature fusion neck, and a parameter-sharing lightweight detection head. Second, unimportant branches of the network are pruned to further compress the model. Finally, the detection accuracy of the model is enhanced by knowledge distillation without augmenting the model volume, which compensates for the accuracy loss caused by model compression. Experimental validation is conducted on three SAR image ship detection datasets (SSDD, high-resolution SAR images dataset, large-scale SAR ship detection dataset-v1.0) to thoroughly assess the effectiveness of the proposed lightweight algorithm. Experimental results on the three datasets demonstrate that the proposed method achieves a model volume reduction to one-third of the baseline while maintaining a minimal decrease in detection accuracy. In SSDD, the proposed method achieved 98.7 accuracy, 0.92M parameters, 3.1G FLOPS and 2.1 MB size of 1.5X pruning rate. Furthermore, it outperforms other state-of-the-art lightweight detectors.

关键词： Computational modeling Accuracy Convolution Marine vehicles Synthetic aperture radar Solid modeling Feature extraction Detectors Adaptation models Remote sensing Convolutional neural network (CNN) knowledge distillation (KD) lightweight network pruning synthetic aperture radar (SAR) target detection

来源：评论

学校读者我要写书评

暂无评论

Near-Field High-Speed User Sensing in Wideband mmWave Communications: Algorithms and Bounds

引用

IEEE TRANSACTIONS ON SIGNAL PROCESSING 2025年 73卷 919-935页

作者： Miao, Hongxia Peng, Mugen Beijing Univ Posts & Telecommun Sch Informat & Commun Engn State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

Integrated sensing and communications (ISAC) has been expected to be a key technique in the sixth-generation cellular networks. With the increase of carrier frequency (to millimeter-wave or Terahertz spectrum) and antenna array size (to extremely large-scale antenna) in wireless communications, the near-field area is enlarged and cannot be ignored. Accordingly, the channel model and its estimation algorithms are changed, which bring new chances in ISAC. However, the impact of both Doppler and spatial wideband effects, caused by high mobility and multicarriers, on sensing performance using communication signals is not well studied. In this study, these two effects are shown to be helpful in user sensing. First, the channel model is proposed for a high-speed moving user transmitting an orthogonal frequency division multiplex (OFDM) signal, where there are six unknown parameters. Then, the Cramer-Rao lower bounds (CRLB) for joint six parameter estimation is determined, where the impact of the near-field parameter and the velocity on the CRLB of positioning are discussed and quantified. Further, to compensate for the deficiency that the CRLB is tight only in high signal-to-noise-ratio (SNR) scenarios, we derive the Ziv-Zakai bound (ZZB) for positioning by exploiting the prior information on positioning parameters. Subsequently, a joint position and velocity parameter estimation algorithm is designed by first performing a discrete fractional Fourier transform on the received signal to obtain a coarse estimation and then refining it by Newton-based refinement. Numerical results coincide with our analysis.

关键词： OFDM Estimation Signal processing algorithms Channel estimation Chirp Frequency estimation Training Signal to noise ratio Sensors Matching pursuit algorithms Near-field ISAC high-speed moving CRLB ZZB

来源：评论

学校读者我要写书评

暂无评论

Differentially private generative decomposed adversarial network for vertically partitioned data sharing

引用

INFORMATION SCIENCES 2023年 619卷 722-744页

作者： Wang, Zhenya Cheng, Xiang Su, Sen Wang, Guangsheng Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

This paper considers the problem of differentially private vertically partitioned data shar-ing. In particular, with the assistance of a semi-honest curator, the involved parties (i.e., data owners) each holds records regarding different features of the same set of individuals collectively generate a shared dataset while satisfying differential privacy. Motivated by the superiority of the generative adversarial network (GAN) in data synthesizing, we pre-sent a differentially private generative decomposed adversarial network (DPGDAN) approach for vertically partitioned data sharing. In DPGDAN, the discriminator in the initial GAN is decomposed into several local discriminators and two relational discriminators, i.e., a real relational discriminator and a fake relational discriminator. Each party is assigned with a local discriminator while the curator holds the generator and the two relational dis-criminators. By combining the sanitized feedback from the local discriminators and the outputs of the relational discriminators, the curator can update the generator to approxi-mate the distribution of the integrated dataset without compromising each party's privacy. Moreover, to promote the shared data's utility, we design a snapshot aggregation method by aggregating the synthetic records produced by the generators acquired during training. Furthermore, we show that DPGDAN satisfies (epsilon, delta)- differential privacy. Extensive experi-ments validate the effectiveness of DPGDAN.

关键词： Differential privacy Multiple parties Data sharing Vertically partitioned data Generative adversarial network

来源：评论

学校读者我要写书评

暂无评论

Classification-labeled Continuousization and Multi-Domain Spatio-Temporal Fusion for Fine-Grained Urban Crime Prediction

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2023年第7期35卷 6725-6738页

作者： Zhao, Shuai Liu, Ruiqiang Cheng, Bo Zhao, Daxing Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Beijing 100876 Peoples R China

Fine-grained urban crime prediction is of great significance to urban management and public safety. Previous crime prediction work has been done at a relatively coarse time granularity, which may suffer from two issues for fine-grained crime prediction. 1) The zero-inflation problem associated with fine-grained granularity. Crime occurrence is sparse, and when the time granularity becomes finer, it leads to a more sparse prediction label for this problem resulting in the zero inflation problem. 2) Insufficient amount of information involved in crime datasets. When the spatio-temporal granularity becomes smaller, more information from related fields needs to be introduced to extract spatio-temporal features to assist the analysis. To address the first issue, we introduce a classification-labeled continuousization strategy and a weighted loss function for sparse classification problem, making the model more likely to focus on non-zero elements in zero-inflated datasets. For the second issue, we propose a novel deep learning based model, termed attention-based spatio-temporal multi-domain fusion network, which fuses features from multiple datasets in related domains. We evaluate our method on six real-world datasets collected in New York City and experiments on our model show the advantages beyond many competitive baselines.

关键词： Spatiotemporal data mining urban computing deep learning

来源：评论

学校读者我要写书评

暂无评论

A novel synchrosqueezing transform associated with linear canonical transform

引用

SIGNAL PROCESSING 2025年 227卷

作者： Miao, Hongxia Beijing Univ Posts & Telecommun Sch Informat & Commun Engn State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

Synchrosqueezing transforms have aroused great attention for its ability in time-frequency energy rearranging and signal reconstruction, which are post-processing techniques of the time-frequency distribution. However, the time-frequency distributions, such as short-time Fourier transform and short-time fractional Fourier transform, cannot change the shape of the time-frequency distribution. The linear canonical transform (LCT) can simultaneously rotate and scale the time-frequency distribution, which enlarges the distance between different signal components with proper parameters. In this study, a convolution-type short-time LCT is proposed to present the time-frequency distribution of a signal, from which the signal reconstruction formula is given. Its resolutions in time and LCT domains are demonstrated, which helps to select suitable window functions. A fast implementation algorithm for the short-time LCT is provided. Further, the synchrosqueezing LCT (SLCT) transform is designed by performing synchrosqueezing technique on the short-time LCT. The SLCT inherits many properties of the LCT, and the signal reconstruction formula is obtained from the SLCT. Adaptive selections of the parameter matrix of LCT and the length of the window function are introduced, thereby enabling proper compress direction and resolution of the signal. Finally, numerical experiments are presented to verify the efficiency of the SLCT.

关键词： Synchrosqueezing transform Generalized convolution operation LCT Chirp Linear time-varying system

来源：评论

学校读者我要写书评

暂无评论

FAT: Field-Aware Transformer for Point Cloud Segmentation With Adaptive Attention Fields

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2024年第9期20卷 10825-10835页

作者： Zhou, Junjie Liu, Baolin Xiong, Yongping Chiu, Chinwai Liu, Fangyu Gong, Xiangyang Beijing Univ Posts & Telecommun Sch Comp Sci State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

Point cloud segmentation is crucial for various industrial applications, such as autonomous driving and robotics. Recent developments underscore the significant potential of transformer models in this field. However, existing attention mechanisms apply the same feature learning paradigm for all points equally, ignoring the considerable size differences among objects in a scene. To rectify this, we introduce the field-aware transformer (FAT), engineered to tailor effective receptive fields to objects of varying sizes. Our FAT achieves field-aware learning through two primary components: the multigranularity attention (MGA) scheme and the reattention module. The MGA scheme is proficient in aggregating tokens from distant areas while preserving multiscale features within each attention layer. The reattention module dynamically adjusts the attention scores to the fine- and coarse-grained features output by MGA for each point. Extensive experimental results underscore the effectiveness and efficiency of our FAT, which delivers state-of-the-art performance on both the stanford 3D indoor scene dataset (S3DIS) and ScanNetV2 datasets.

关键词： 3-D scene understanding multigranularity attention (MGA) point cloud semantic segmentation transformer model

来源：评论

学校读者我要写书评

暂无评论

Quantum private query protocol based on counterfactual quantum key distribution with noiseless attack

引用

QUANTUM INFORMATION PROCESSING 2024年第10期23卷 1-15页

作者： Liu, Dongmei Li, Jian Chen, Xiubo Ye, Chongqiang Wang, Zhuo Beijing Univ Posts & Telecommun Informat Secur Ctr State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

Most of the previous quantum private query protocols are based on the BB84 key distribution type and generally use a method that wastes most quantum resources to complete security detection. In order to save resources, this paper proposes a noiseless counterfactual quantum private query protocol with high efficiency. This protocol improves the counterfactual quantum key distribution protocol proposed by Rao and Srikanth (Phys Rev A 104:022424, 2021. https://***/10.1103/PhysRevA.104.022424). The communicating parties randomly select quantum bits with probability f to perform flipping to complete the quantum private query. The detector at the sending end receives the counterfactual bit, which we use for security detection;the detector at the receiving end is called a non-counterfactual bit, which is combined with random bit flipping to realize key transmission, followed by traditional post-processing and private query. Finally, when f=0.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f=0.5$$\end{document}, the response probability (keys rate) of the non-counterfactual detector at the receiving end can reach a minimum value of 0.5.

关键词： Counterfactual quantum key distribution Quantum private query High efficiency Noiseless

来源：评论

学校读者我要写书评

暂无评论

MVC-RSN: A Malware Classification Method With Variant Identification Ability

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第21期11卷 35654-35668页

作者： Wu, Wei Peng, Haipeng Zhu, Haotian Li, Lixiang Beijing Univ Posts & Telecommun Informat Secur Ctr State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

With the rapid development of the Internet of Things (IoT), a substantial number of mobile devices may need to be connected to and communicate with the other devices. IoT smart nodes, which play a key role in these processes, provide not only connectivity but also data security during data collection and transmission. While existing research has considered security issues concerning data collection and transmission to some extent, the inherent vulnerabilities of smart nodes themselves have rarely received adequate attention. Actually, smart nodes can be susceptible to malware attacks, and the fundamental task to protect them is the timely and accurate identification of malware. However, the stealthy nature of malware, the imbalance in its types, and the sharp increase in malware variants make this task challenging. Existing static malware classification methods, which require known malware features, may struggle with a whole new set of malware variants. Dynamic methods plausibly have the potential to identify malware variants but necessitate running malware beforehand. In this article, we propose a deep-learning-based malware classification method designed to help most smart IoT nodes running mainstream operating systems preventing attacks from the malware and its variants. The proposed method incorporates an automatically adjusting threshold mechanism to extract trivial features of malware, refining the classification process even if original features are modified. It also performs well with insufficient malware samples and is compatible with most mainstream convolutional neural network (CNN) models allowing flexible implements. Experiments concerning byte-level variants, operation-code-level variants, and noise-caused variants are conducted. The results show that for the malware files and their variants, the proposed method can accurately accomplish the malware classification tasks. In addition, the proposed method offers a robust solution for the malware classifi

关键词： deep learning malware classification Convolutional neural network (CNN) variant variant malware classification variant

来源：评论

学校读者我要写书评

暂无评论

Flight Trajectory Control With Network-Oriented Hierarchical Reinforcement Learning for UAVs-Assisted Data Time-Sensitive IoT

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2025年第5期26卷 6332-6345页

作者： Wang, Jingyu Li, Songyang Chen, Dezhi Fu, Qianlong Qi, Qi Sun, Haifeng He, Bo Liao, Jianxin Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China Pengcheng Lab Shenzhen 518000 Peoples R China

Within the Internet of Things (IoT) for traffic monitoring, the employment of autonomous aerial vehicles (AAVs) as relays for collecting and transmitting real-time data from traffic sensors to base stations has proven a promising approach. In UAV-assisted Data Time-Sensitive IoT (DTIoT), the Age of Information is a crucial metric assessing data freshness, measuring the elapsed time from traffic sensors to the base station. Optimizing flight trajectories of multiple UAVs to minimize AoI while adhering to energy constraints poses a significant challenge. Current research often employs deep reinforcement learning for UAV trajectory control. Nevertheless, managing multi-agent continuous trajectories in intricate DTIoT network environments faces obstacles due to sparse rewards, thus impeding the training of deep neural network-based control policies using traditional DRL techniques. In this paper, we propose a network-oriented hierarchical reinforcement learning (NO-HRL) to control the UAVs' flight trajectory in DTIoT networks for minimizing the AoI. We devise a control policy leveraging a two-tier hierarchical DRL framework, with the upper tier determining the target and the lower tier executing it. We also introduce a decoupled sequential training approach to efficiently train the mutually dependent two-tier DRL network of NO-HRL. Experimental results demonstrate that our method excels in optimizing AoI for DTIoT compared to other baselines.

关键词： Trajectory Autonomous aerial vehicles Sensors Training Internet of Things Aerospace electronics Monitoring Data collection Base stations Information age Autonomous aerial vehicle (AAV) sensor data collection age of information hierarchical reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

FPCA: Parasitic Coding Authentication for UAVs by FM Signals

引用

IEEE-ACM TRANSACTIONS ON networking 2024年第6期32卷 4570-4584页

作者： Zhu, Shaopeng Zheng, Xiaolong Liu, Liang Ma, Huadong Beijing Univ Posts & Telecommun Sch Comp Sci State Key Lab Networking & Switching Technol Beijing 100088 Peoples R China

De-authentication attack is one of the major threats to Unmanned Aerial Vehicle (UAV) communication, in which the attacker continuously sends de-authentication frames to disconnect the UAV communication link. Existing defense methods are based on authentication by digital passwords or physical channel features. But they suffer from replay attacks or cannot adapt to the UAV mobility. In this paper, instead of enhancing the in-channel authentication, we leverage the ambient broadcasting signal to establish a low-cost additional channel for authentication. Different from methods using another dedicated secure communication channel to perform an independent authentication, we use the ambient FM radio broadcasting channel and couple the two channels by encoding parasitic bits on the host signals of the broadcasting channel, which is called parasitic coding. To further enhance the security, we propose the FM-based Parasitic Coding Authentication (FPCA) that leverages elaborate host signal processing and vector coding to ensure that the attacker cannot decode our authentication even knowing the FM receiving frequency. We implement FPCA on the embedded UAV platform. The extensive experiments show that FPCA can resist replay attacks and brute force searching, achieving reliable continuous authentication for UAVs.

关键词： Authentication Autonomous aerial vehicles Encoding Frequency modulation Reliability Receivers Security Broadcasting Wireless communication Vectors UAV de-authentication attack authentication FM signal vector coding

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：