检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Cheng, Qiankun Dong, Rongen Cai, Wenlong Liu, Ruiqi Shu, Feng Wang, Jiangzhou The School of Information and Communication Engineering Hainan University Haikou570228 China The School of Information and Communication Engineering Collaborative Innovation Center of Information Technology Hainan University Haikou570228 China The School of Electronic and Optical Engineering Nanjing University of Science and Technology Nanjing210094 China National Key Laboratory of Science and Technology on Aerospace Intelligence Control Beijing Aerospace Automatic Control Institute Beijing China The Wireless and Computing Research Institute ZTE Corporation Beijing100029 China The School of Engineering University of Kent CanterburyCT2 7NT United Kingdom

Due to its ability of overcoming the impact of double-fading effect, active intelligent reflecting surface (IRS) has attracted a lot of attention. Unlike passive IRS, active IRS should be supplied by power, thus adjusting power between base station (BS) and IRS having a direct impact on the system rate performance. In this paper, the active IRS-aided network under a total power constraint is modeled with an ability of adjusting power between BS and IRS. Given the transmit beamforming at BS and reflecting beamforming at IRS, the SNR expression is derived to be a function of power allocation (PA) factor, and the optimization of maximizing the SNR is given. Subsequently, two high-performance PA strategies, enhanced multiple random initialization Newton's (EMRIN) and Taylor polynomial approximation (TPA), are proposed. The former is to improve the rate performance of classic Netwon's method to avoid involving a local optimal point by using multiple random initializations. To reduce its high computational complexity, the latter provides a closed-form solution by making use of the first-order Taylor polynomial approximation to the original SNR function. Actually, using TPA, the original optimization problem is transformed into a problem of finding a root for a third-order polynomial. Simulation results are as follows: the first-order TPA of SNR fit its exact expression well, the proposed two PA methods performs much better than fixed PA in accordance with rate, and appoaches exhaustive search as the number of IRS reflecting elements goes to large-scale. Copyright © 2023, The Authors. All rights reserved.

关键词： Polynomial approximation

来源：评论

学校读者我要写书评

暂无评论

Coalition Formation for Heterogeneous Federated Learning Enabled Channel Estimation in RIS-assisted Cell-free MIMO

arXiv

引用

arXiv 2025年

作者： Qi, Nan Liu, Haoxuan Tsiftsis, Theodoros A. Boulogeorgos, Alexandros-Apostolos A. Zhou, Fuhui Jin, Shi Wu, Qihui Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space Ministry of Industry and Information Technology Nanjing University of Aeronautics and Astronautics Nanjing210016 China Department of Informatics and Telecommunications University of Thessaly Lamia35100 Greece Department of Electrical and Computer Engineering University of Western Macedonia Kozani50100 Greece National Mobile Communications Research Laboratory Frontiers Science Center for Mobile Information Communication and Security Southeast University Nanjing210096 China

Downlink channel estimation remains a significant bottleneck in reconfigurable intelligent surface-assisted cell-free multiple-input multiple-output communication systems. Conventional approaches primarily rely on centralized deep learning methods to estimate the high-dimensional and complex cascaded channels. These methods require data aggregation from all users for centralized model training, leading to excessive communication overhead and significant data privacy concerns. Additionally, the large size of local learning models imposes heavy computational demands on end users, necessitating strong computational capabilities that most commercial devices lack. To address the aforementioned challenges, a coalition-formation-guided heterogeneous federated learning (FL) framework is proposed. This framework leverages coalition formation to guide the formation of heterogeneous FL user groups for efficient channel estimation. Specifically, by utilizing a distributed deep reinforcement learning (DRL) approach, each FL user intelligently and independently decides whether to join or leave a coalition, aiming at improving channel estimation accuracy, while reducing local model size and computational costs for end users. Moreover, to accelerate the DRL-FL convergence process and reduce computational burdens on end users, a transfer learning method is introduced. This method incorporates both received reference signal power and distance similarity metrics, by considering that nodes with similar distances to the base station and comparable received signal power have a strong likelihood of experiencing similar channel fading. Massive experiments performed that reveal that, compared with the benchmarks, the proposed framework significantly reduces the computational overhead of end users by 16%, improves data privacy, and improves channel estimation accuracy by 20%. © 2025, CC BY.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Downstream-agnostic Adversarial Examples

arXiv

引用

arXiv 2023年

作者： Zhou, Ziqi Hu, Shengshan Zhao, Ruizhi Wang, Qian Zhang, Leo Yu Hou, Junhui Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Cyber Science and Engineering Wuhan University China School of Information and Communication Technology Griffith University Australia Department of Computer Science City University of Hong Kong Hong Kong National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning operations to enjoy the benefit of "large model". Despite this promising prospect, the security of pre-trained encoder has not been thoroughly investigated yet, especially when the pre-trained encoder is publicly available for commercial use. In this paper, we propose AdvEncoder, the first framework for generating downstream-agnostic universal adversarial examples based on the pre-trained encoder. AdvEncoder aims to construct a universal adversarial perturbation or patch for a set of natural images that can fool all the downstream tasks inheriting the victim pre-trained encoder. Unlike traditional adversarial example works, the pre-trained encoder only outputs feature vectors rather than classification labels. Therefore, we first exploit the high frequency component information of the image to guide the generation of adversarial examples. Then we design a generative attack framework to construct adversarial perturbations/patches by learning the distribution of the attack surrogate dataset to improve their attack success rates and transferability. Our results show that an attacker can successfully attack downstream tasks without knowing either the pre-training dataset or the downstream dataset. We also tailor four defenses for pre-trained encoders, the results of which further prove the attack ability of AdvEncoder. Our codes are available at: https://github. com/CGCL-codes/AdvEncoder. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Multi-BS Fusion Scheme for PHD Simultaneous Localization and Mapping

Multi-BS Fusion Scheme for PHD Simultaneous Localization and...

引用

International Conference on communication technology (ICCT)

作者： Tao Du Jie Yang Shuqiang Xia Shi Jin National Mobile Communications Research Laboratory Southeast University Nanjing Jiangsu China Frontiers Science Center for Mobile Information Communication and Security Southeast University Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Southeast University Nanjing China ZTE Corporation Shenzhen China State Key Laboratory of Mobile Network and Mobile Multimedia Shenzhen China

Integrated sensing and communication (ISAC) is expected to improve the energy and spectral efficiency of communication systems. Probability hypothesis density (PHD) simultaneous localization and mapping (SLAM) without explicit data association is one of the mainstream algorithms for implementing sensing function in ISAC. However, most existing research focused on multi-sensor and multi-user sensing fusion, while few studies involve multi base station (BS) fusion schemes. In this study, we introduce the virtual reference point rather than common used virtual anchor to model reflecting surface to form a consistent description of the surrounding environment among different BSs. Then, the estimation results of BSs are sequentially utilized as prior information, which is compared with arithmetic average (AA) and generalized covariance intersection (GCI) fusion algorithms. To improve the poor performance of existing schemes when filed of view (FoV) of different BSs is partial overlapped, we propose the concept of establishing global cloud map. The AA fusion algorithm is integrated into cloud fusion to improve the map estimation performance. Simulation results demonstrate that the performance of sequential fusion is superior to AA and GCI schemes when FoVs of different BSs are fully overlapped. Besides, the cloud map fusion scheme can obtain global maps even if some features are no longer in FoV, as well as improve the accuracy and stability of local maps estimation for each BS.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Remote Attestation for Intelligent Electronic Devices in Smart Grid Based on Trusted Level Measurement

引用

Chinese Journal of Electronics 2020年第3期29卷 437-446页

作者： WANG Yong LI Jun'e CHEN Xiong LIN Hai YU Fajiang LUO Jianbo Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of EducationSchool of Cyber Science and Engineering Wuhan University NARI Group Corporation State Grid Electric Power Research Institute NARI Technology Co.Ltd. State Key Laboratory of Smart Grid Protection and Control

To reduce the implementation complexity and excessive overhead caused by remote attestation,we propose a new remote attestation scheme based on trusted level measurement considering the limited system resources of the Intelligent electronic devices(IEDs) in smart grid. In this scheme, a lightweight trusted level measurement mechanism is designed, and a new dynamic weighted multi-classifier integration method based on Binary tree Support vector machine(BT-SVM) is proposed. The trusted level of configuration information and running state of an IED are measured through trusted third party. The measurement parameters and attestation period are dynamically adjusted according to different trusted levels of an IED, so that reducing computational overhead and communication delays in remote attestation process. Analysis and experiments show that the proposed scheme can enhance the security of IEDs in smart grid by occupying a small amount of computation and communication resources.

关键词： Smart grid Intelligent electronic devices(IEDs) Remote attestation Trusted level measurement

来源：评论

学校读者我要写书评

暂无评论

Automatic Smart Contract Generation with Knowledge Extraction and Unified Modeling Language 7th

Automatic Smart Contract Generation with Knowledge Extracti...

引用

7th International Conference on Smart Computing and communication, SmartCom 2022

作者： Ran, Peiyun Liu, Mingsheng Zheng, Jianwu Bhuiyan, Zakirul Alam Li, Jianhua Li, Gang Yu, Shiyuan Wang, Lifeng Tang, Song Zhao, Peng School of Cyberspace Security of Beihang University Beijing China Shijiazhuang Institute of Railway Technology Hebei Shijiazhuang050041 China Key Laboratory of Traffic Safety and Control of Hebei Province Shijiazhuang China School of Transportation Shijiazhuang Tiedao University Hebei Shaoxing China Department of Computer and Information Sciences Fordham University JMH 334 E Fordham Road BronxNY10458 United States School of Information Science and Technology Shijiazhuang TieDao University No. 17 East North Second Ring Road Changan District Shijiazhuang China Co. Ltd. Beijing China Beijing Public Security Burea Beijing China Hebei Huaye Jike Information Technology Co. Ltd. Shijiazhuang China Institute of Applied Mathematics of Hebei Academy of Sciences Shijiazhuang China The First Hospital of Hebei Medical University Shijiazhuang China

ISBN: (纸本)9783031281235

Since the launch of Ethereum in 2013, the smart contract has been a momentous part of the blockchain systems due to its character of automatic execution. The generation of smart contracts has also attracted extensive attention from the academic community. However, the preparation and generation of smart contracts are still mainly manual so far, which limits the scalability of the smart contracts. In this paper, we put forward a new method to generate smart contracts automatically based on knowledge extraction and Unified Modeling Language (UML), which can significantly accelerate the generation of smart contracts. We will describe this method in more detail based on the logistics supply chain. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Timeliness-Aware Multiple Rumor Sources Estimation in Dynamic Online Social Networks

SSRN

引用

SSRN 2024年

作者： Huang, Da-Wen Wu, Wenjie Bi, Jichao Li, Junli Gan, Chenquan Zhou, Wei College of Computer Science Sichuan Normal University Chengdu China State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou China School of Cyber Security and Information Law Chongqing University of Posts and Telecommunications Chongqing China School of Big Data and Software Engineering Chongqing University Chongqing China

Identifying rumor sources in online social networks (OSNs) plays a crucial role in controlling the spread of rumors and mitigating the damage caused by them. However, most studies are not suitable for identifying rumor sources in dynamic OSNs with community structures. Moreover, these studies tend to neglect the impact of timeliness on the spread of rumors, potentially leading to an overestimation of rumors' propagation ability. To overcome these limitations, this paper proposes a multiple rumor sources estimation framework for OSNs. First, a community-based dynamic network model is introduced to depict the temporal nature and community feature of OSNs. Second, a meticulously designed microcosmic SIR model, incorporating the timeliness of rumor topics, is developed to unravel the complex dynamics of rumor propagation. Then, a computationally efficient multiple rumor sources estimation algorithm is proposed. This algorithm utilizes the infection information collected by sensors and applies the maximum likelihood estimation (MLE) method to identify the rumor sources. Finally, experimental results on real-world and synthetic temporal networks demonstrate the effectiveness of the proposed rumor source estimation algorithm. © 2024, The Authors. All rights reserved.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Laser Induced Nano-Hole for High Throughput Fabrication of Silver Doped Chalcogenide Metasurfaces

SSRN

引用

SSRN 2022年

作者： Zhou, Jianhong Zhao, Jianxing Cao, Lixiao Zhou, Yao Wang, Li Wang, Chonge Jing, Wenbo Zhang, Gong Yu, Ying School of Photoelectric Engineering Changchun University of Science and Technology Changchun130022 China Science and Technology on Electro-Optical Information Security Control Laboratory China Electronics Technology Group Corporation 53rd Research Institute Tianjin300308 China

We report the potential method for high throughput chalcogenide metasurface fabrication by using focused 632.8 nm laser induced nano-hole effect in the silver doped Ge28Sb12Se60 film. Generally, laser induced dewetting effect is hard to realize in thin Ge28Sb12Se60 films, however with the assistance of silver doping, the randomly distributed nano holes can be acquired in the dewetted region. The higher the silver concentration, the easier the laser induced nano-hole effect is, but much higher silver concentration like 47.8% will suppress the laser induced nano-hole effect due to the photocrystallisation. The nano holes can be enlarged and further connect with adjacent ones by increasing the output laser power, thus the morphology of the laser induced nano holes can be flexibly manipulated. Our findings are beneficial in high-throughput fabrication of chalcogenide metamaterial. © 2022, The Authors. All rights reserved.

关键词： Silver

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Feature Enrichment and Fidelity Preservation Learning Framework for Skeleton-based Action Recognition

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年

作者： Li, Chuankun Li, Shuai Gao, Yanbo Gao, Xingyu Chen, Ping Li, Jian Li, Wanqing North University of China State Key Laboratory of Dynamic Testing Technology School of Information and Communication Engineering Taiyuan030051 China Shandong University School of Control Science and Engineering School of Software Jinan250100 China Chinese Academy of Sciences Institute of Microelectronics Beijing100029 China University of Wollongong Advanced Multimedia Research Lab WollongongNSW2522 Australia

Unsupervised skeleton-based action recognition has achieved remarkable progress recently. Existing unsupervised learning methods suffer from severe overfitting problem, and thus small networks are used, significantly reducing the representation capability. To address this problem, the overfitting mechanism behind the unsupervised learning for skeleton-based action recognition is first investigated. It is observed that skeleton is already a relatively high-level and low-dimension feature, but not in the same manifold as the features for action recognition. Simply applying the existing unsupervised learning method tends to produce features that discriminate the different samples rather than action classes, resulting in the overfitting problem. To address this problem, this paper proposes an Unsupervised spatial-temporal Feature Enrichment and Fidelity Preservation (U-FEFP) learning framework to generate rich distributed features that contain all the information of a skeleton sample. A spatial-temporal feature transformation subnetwork is developed using channel-wise topology refinement graph convolutional block and graph convolutional gated recurrent unit block as the basic feature extraction network. The unsupervised Bootstrap Your Own Latent-based learning is utilized to generate rich distributed features, and the unsupervised pretext task-based learning is employed to preserve the information contained in the skeleton. The two unsupervised learning ways are collaborated as U-FEFP to produce robust and discriminative representations. Experimental results on four widely used benchmarks, namely NTU-RGB+D-60, PKU-MMD, NTU-RGB+D-120 and UAV-Human dataset, demonstrate that the proposed U-FEFP obtains the best result compared with the state-of-the-art unsupervised learning methods. © 1991-2012 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

RePAD: Real-Time Proactive Anomaly Detection for Time Series 34th

RePAD: Real-Time Proactive Anomaly Detection for Time Series

引用

34th International Conference on Advanced information Networking and Applications, AINA 2020

作者： Lee, Ming-Chang Lin, Jia-Chun Gran, Ernst Gunnar Department of Information Security and Communication Technology Norwegian University of Science and Technology Gjøvik2815 Norway Simula Research Laboratory Fornebu1364 Norway

ISBN: (纸本)9783030440404

During the past decade, many anomaly detection approaches have been introduced in different fields such as network monitoring, fraud detection, and intrusion detection. However, they require understanding of data pattern and often need a long off-line period to build a model or network for the target data. Providing real-time and proactive anomaly detection for streaming time series without human intervention and domain knowledge is highly valuable since it greatly reduces human effort and enables appropriate countermeasures to be undertaken before a disastrous damage, failure, or other harmful event occurs. However, this issue has not been well studied yet. To address it, this paper proposes RePAD, which is a Real-time Proactive Anomaly Detection algorithm for streaming time series based on Long Short-Term Memory (LSTM). RePAD utilizes short-term historical data points to predict and determine whether or not the upcoming data point is a sign that an anomaly is likely to happen in the near future. By dynamically adjusting the detection threshold over time, RePAD is able to tolerate minor pattern change in time series and detect anomalies either proactively or on time. Experiments based on two time series datasets collected from the Numenta Anomaly Benchmark demonstrate that RePAD is able to proactively detect anomalies and provide early warnings in real time without human intervention and domain knowledge. © 2020, Springer Nature Switzerland AG.

关键词： Time series

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：