检索结果-内蒙古大学图书馆

A Graph-Based Semi-Supervised Approach for Few-Shot Class-Incremental Modulation Classification

China communications 2024年第11期21卷 88-103页

作者： Zhou Xiaoyu Qi Peihan Liu Qi Ding Yuanlei Zheng Shilian Li Zan State Key Laboratory of Integrated Service Networks Xidian UniversityXi’an 710071China Xinjiang Astronomical Observatory Chinese Academy of SciencesUrumqi 830011China The No.011 Research Center Science and Technology on Communication Information Security Control LaboratoryJiaxing 314033China

With the successive application of deep learning(DL)in classification tasks,the DL-based modulation classification method has become the preference for its state-of-the-art ***,once the DL recognition model is pre-trained with fixed classes,the pre-trained model tends to predict incorrect results when identifying incremental ***,the incremental classes are usually emergent without label information or only a few labeled samples of incremental classes can be *** this context,we propose a graphbased semi-supervised approach to address the fewshot classes-incremental(FSCI)modulation classification *** proposed method is a twostage learning method,specifically,a warm-up model is trained for classifying old classes and incremental classes,where the unlabeled samples of incremental classes are uniformly labeled with the same label to alleviate the damage of the class imbalance *** the warm-up model is regarded as a feature extractor for constructing a similar graph to connect labeled samples and unlabeled samples,and the label propagation algorithm is adopted to propagate the label information from labeled nodes to unlabeled nodes in the graph to achieve the purpose of incremental classes *** results prove that the proposed method is superior to other finetuning methods and retrain methods.

关键词： deep learning few-shot label propagation modulation classification semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Model-based deep learning for fiber bundle infrared image restoration

引用

Defence technology（防务技术） 2023年第9期27卷 38-45页

作者： Bo-wen Wang Le Li Hai-bo Yang Jia-xin Chen Yu-hai Li Qian Chen Chao Zuo Jiangsu Key Laboratory of Spectral Imaging and Intelligent Sense Nanjing University of Science and TechnologyNanjingJiangsu Province210094China Smart Computational Imaging(SCI)Laboratory Nanjing University of Science and TechnologyNanjingJiangsu Province210094China Science and Technology on Electro-Optical Information Security Control Lab Tianjin300308China

As the representative of flexibility in optical imaging media,in recent years,fiber bundles have emerged as a promising architecture in the development of compact visual *** to tackling the problems of universal honeycomb artifacts and low signal-to-noise ratio(SNR)imaging in fiber bundles,the iterative super-resolution reconstruction network based on a physical model is *** the constraint of solving the two subproblems of data fidelity and prior regularization term alternately,the network can efficiently“regenerate”the lost spatial resolution with deep *** building and calibrating a dual-path imaging system,the real-world dataset where paired low-resolution(LR)-high-resolution(HR)images on the same scene can be generated *** results on both the United States Air Force(USAF)resolution target and complex target objects demonstrate that the algorithm can restore high-contrast images without pixilated *** the basis of super-resolution reconstruction,compound eye image composition based on fiber bundle is also embedded in this paper for the actual imaging *** proposed work is the first to apply a physical model-based deep learning network to fiber bundle imaging in the infrared band,effectively promoting the engineering application of thermal radiation detection.

关键词： Fiber bundle Deep learning Infrared imaging Image restoration

来源：评论

学校读者我要写书评

暂无评论

BotCapturer: Detecting botnets based on two-layered analysis with graph anomaly detection and network traffic clustering

引用

International Journal of Performability Engineering 2018年第5期14卷 1050-1059页

作者： Wang, Wei Wang, Yang Tan, Xinlu Liu, Ya Yang, Shuangmao Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing Jiaotong University Beijing100044 China Science and Technology on Electronic Information Control Laboratory Chengdu610036 China

Botnets have become one of the most serious threats on the Internet. On the platform of botnets, attackers conduct series of malicious activities such as distributed denial-of-service (DDoS) or virtual currencies mining. Network traffic has been widely used as the data source for the detection of botnets. However, there are two main issues on the detection of botnets with network traffic. First, many traditional filtering methods such as whitelisting are not able to process the very large amount of traffic data in real-time due to their limited computational capability. Second, many existing detection methods, based on network traffic clustering, result in high false positive rates. In this work, we are motivated to resolve the above two issues by proposing a lightweight botnet detection system called BotCapturer, based on two-layered analysis with anomaly detection in graph and network communication traffic clustering. First, we identify anomalous nodes that correspond to C&C (control and Command) servers with anomaly scores in a graph abstracted from the network traffic. Second, we take advantage of clustering algorithms to check whether the nodes interacting with an anomalous node share similar communication pattern. In order to minimize irrelevant traffic, we propose a traffic reduction method to reduce more than 85% background traffic. The reduction is conducted by filtering the packets that are unrelated to the hosts like C&C server. We collect a very big dataset by simulating five different botnets and mixing the collected traffic with background traffic obtained from ISP. Extensive experiments are conducted and evaluation results based on our own dataset show that BotCapturer reduces more than 85% input raw packet traces and achieves a high detection rate (100%) with a low false positive rate (0.01%), demonstrating that it is very effective and efficient in detecting latest botnets. © 2018 Totem Publisher, Inc. All rights reserved.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Optimization of the Internet of Remote Things Data Acquisition Based on Satellite UAV Integrated Network

引用

China communications 2023年第7期20卷 15-28页

作者： Yuanyuan Yao Dengyang Dong Sai Huang Chunyu Pan Shuo Chen Xuehua Li Key Laboratory of Information and Communication Systems Ministry of Information IndustryBeijing Information Science and Technology UniversityBeijing 100101China Key Laboratory of Modern Measurement&Control Technology Ministry of EducationBeijing Information Science and Technology UniversityBeijing 100101China Key Laboratory of Universal Wireless Communications Ministry of EducationBeijing University of Posts and TelecommunicationsBeijing 100876China

In order to achieve dependable and efficient data acquisition and transmission in the Internet of Remote Things(IoRT),we investigate the optimization scheme of IoRT data acquisition under the unmanned aerial vehicle(UAV)-low earth orbit(LEO)satellite integrated space-air-ground network,in which the UAV acquires data from massive Internet of Things(IoT)devices in special *** combine with the actual scenario,we consider two different data types,that is,delay-sensitive data and delay-tolerant data,the transmission mode is accordingly divided into two *** delay-sensitive data,the data will be transmitted via the LEO satellite relay to the data center(DC)in *** delay-tolerant data,the UAV will store and carry the data until the acquisition is completed,and then return to *** to nonconvexity and complexity of the formulated problem,a multi-dimensional optimization Rate Demand based Joint Optimization(RDJO)algorithm is *** algorithm first uses successive convex approximation(SCA)technology to solve the non-convexity,and then based on the block coordinate descent(BCD)method,the data acquisition efficiency is maximized by jointly optimizing UAV deployment,the bandwidth allocation of IoRT devices,and the transmission power of the ***,the proposed RDJO algorithm is compared with the conventional *** consequences demonstrate that the efficiency of IoRT data acquisition can be greatly improved by multi-parameter optimization of the bandwidth allocation,UAV deployment and the transmission power.

关键词： Internet of Remote Things dataacquisi-tion unmanned aerial vehicle LEO satellite

来源：评论

学校读者我要写书评

暂无评论

Congestion control of Multi-layer Cellular Networks based on Modeling of Transmit Power 3

Congestion Control of Multi-layer Cellular Networks based on...

引用

2020 3rd International Conference on Modeling, Simulation and Optimization Technologies and Applications, MSOTA 2020

作者： Dai, Jiahao Chen, Yongqun Chen, Yucheng Meng, Aodong Science and Technology on Communication Information Security Control Laboratory China China Electronic Group Corporation NO.36 Institute Zhejiang Jiaxing China

The explosive growth of network traffic and the limited storage node resources make the existing network architecture bear great pressure, and network congestion appears in work and life from time to time. In order to ensure the timeliness and effectiveness of the network, we propose a congestion control method based on transmission power. Different from previous work, we built a multi-layer cellular network consisting of multiple macro stations, small base stations and mobile terminals. On this basis, we establish a congestion control model based on the mixed queuing model of service window and calculate the input/output rate and congestion. Finally, we use simulation to verify the effectiveness of the congestion model. The network builder can use our model to set the appropriate power of macro stations and small base stations to ensure network stability. © Published under licence by IOP Publishing Ltd.

关键词： Base stations

来源：评论

学校读者我要写书评

暂无评论

Defense of Signal Modulation Classification Attack Based on GAN 7th

Defense of Signal Modulation Classification Attack Based on ...

引用

7th China National Conference on Big Data and Social Computing, BDSC 2022

作者： Zhu, Jiawei Liu, Yuhang Xu, Dongwei Yang, Hao Xuan, Qi Wang, Shunling Science and Technology on Communication Information Security Control Laboratory Jiaxing314033 China Institute of Cyberspace Security Zhejiang University of Technology Hangzhou310023 China College of Information Engineering Zhejiang University of Technology Hangzhou310023 China

ISBN: (纸本)9789811975318

Although deep learning has been broadly used in all kinds of domains and has also achieved very good performance, it is obvious that deep learning is very vulnerable to be attacked shown by extensive researches (e.g. even a very small perturbation is added to the raw signal data, the result of the modulation classification of the signal by the deep neural network will be wrong), which will cause serious consequences. Thus, it is necessary to find some ways to defense this attack. As a new deep learning model in recent years, Generative Adversarial Network (GAN) has been applied to the field of radio communication, which can encrypt the signal data, so as to ensure the reliable transmission of the signal. In this paper, a defense method against signal modulation classification attack based on GAN is developed, which is used to utilize generator to reconstruct the signal sample to improve the classification accuracy of the model, while the defense ability of the model against adversarial samples is improved. It has been shown that the method achieved good results in experiments. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on performance analysis and optimization methods of bistatic collaborative detection 13

Research on performance analysis and optimization methods of...

引用

13th ACM International Conference on Underwater Networks and Systems, WUWNet 2018

作者： Guo, Longxiang Zhou, Qing Liu, Ting Zou, Jiayun Sheng, Xueli Science and Technology on Communication Information Security Control Laboratory JiaxingZhejiang China College of Underwater Acoustic Engineering Acoustic Science and Technology Laboratory Harbin Engineering University HarbinHeilongjiang China

ISBN: (纸本)9781450361934

Compared with the monostatic sonar system, the bistatic sonar system has numerous advantages, such as being quick and flexible to deploy, utilizing strong concealment, and having a large detection range. The detection performance analysis system, which consists of the sound source and propagation loss level, the interfering noise level, the marine reverberation average strength level, and the target echo sound level, was established in this study. Aiming at a specified area, this study realizes a performance analysis based on the detection probabilities of targets in different locations. In terms of detection performance, the bistatic sonar system especially faces considerable direct wave interference. Therefore, a method for restraining direct wave interference based on zero constraint beam-forming was developed. This method can mitigate the both narrow-band and broadband direct wave interference problems. Another method based on space timeadaptive processingwas also developed. This method can also solve direct wave interference problem. Using the above performance analysis system, considering the influence of strong direct wave interference, the detection range can increase 35% by using the zero constraint beam-forming method, thus further enhancing the detection performance. © 2018 Association for Computing Machinery.

关键词： Sonar

来源：评论

学校读者我要写书评

暂无评论

Semi-fragile watermarking for image authentication based on compressive sensing

引用

science China(information sciences) 2016年第5期59卷 248-250页

作者： Ling DU Xiaochun CAO Wei ZHANG Xinpeng ZHANG Na LIU Jianguo WEI School of Computer Science and Technology Tianjin University School of Computer Shenyang Aerospace University State Key Laboratory of Information Security Institute of Information EngineeringChinese Academy of Sciences School of Communication and Information Engineering Shanghai University

Dear editor,With the aid of sophisticated photoediting software,multimedia content security is becoming increasingly prominent[1,2].The problems of verifying the integrity and recovering the alterations have been a major research topic for information security and multimedia[3–6].Semi-fragile watermarking techniques are more powerful in dealing with such cases,which are fragile to malicious modifications meanwhile robust to incidental

关键词： DCT Semi-fragile watermarking for image authentication based on compressive sensing

来源：评论

学校读者我要写书评

暂无评论

Metamaterial analogue of electromagnetically induced transparency

Metamaterial analogue of electromagnetically induced transpa...

引用

2012 5th Global Symposium on Millimeter-Waves, GSMM 2012

作者： Hua, Jun Chen, Mei Liang Zhu, Lei Meng, Fan Yi Wu, Qun Science and Technology on Communication Information Security Control Laboratory Jiaxing 314033 China Dept. of Microwave Engineering Harbin Institute of Technology Harbin 150001 China

ISBN: (纸本)9781467313032

In this paper, three types of EIT metamaterials are presented and numerically investigated. The type I metamaterial consists of the SR and the SRR, which are coupled with each other and assigned as the dark resonator and the radiative resonator, respectively. The metamaterial can exhibit EIT effect to both normally and laterally incident electromagnetic waves because of its electric and magnetic response, respectively. The type II magnetic metamaterial consists of the split ring resonator (SRR) and the spiral resonator (SR) along two orthogonal axes. For the excitation with the magnetic field perpendicular to the SRR plane, the SRR and the SR are designated as the radiative element and the dark element, respectively. A transparency window is observed in the overlapped absorption band of the independent resonator because of the destructive interference between the radiative and dark resonators of the metamaterial. The type III metamaterial is consisted of three cut wires of different sizes and parallel to each other. The resonance frequencies are different because the lengths of three metal wires are different. Under the polarization of electrical field, because three cut wires experience strong resonance, two-two overlaps of absorption bands of three cut wires will induce two transparency windows. Numerical simulations are carried out to validate the three types of EIT metamaterials, and the results have proven to be quite effective. © 2012 IEEE.

关键词： Metamaterials

来源：评论

学校读者我要写书评

暂无评论

High gain patch antenna with broadband metamaterial lens

High gain patch antenna with broadband metamaterial lens

引用

2012 5th Global Symposium on Millimeter-Waves, GSMM 2012

作者： Hua, Jun Huang, Shuai Yang, Cheng Wu, Qun Science and Technology on Communication Information Security Control Laboratory Jiaxing 314033 China Dept. of Microwave Engineering Harbin Institute of Technology Harbin 150001 China

ISBN: (纸本)9781467313032

Antenna lens can obviously eincrease the gain of the antenna by method of gathering. Also, metamaterial lens can achieve the gathering effect with aspheric surface. In this paper, a new type of metamaterial lens is designed. It is easy to identify as Left-hand material with double negative electromagnetic constant, realizing a negative permittivity and a negative permeability simultaneously. The new structure can perform at a lower frequency and the band of the structure can get a good result in broad band. In addition, the new proposed metamaterial lens do not change the S-parameter of the original antenna. The gain can be increased by 2 dB with the new structure of the metamaterial lens. © 2012 IEEE.

关键词： Microstrip antennas

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：