检索结果-内蒙古大学图书馆

DynaMo: Dynamic Community detection by Incrementally Maximizing Modularity

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2021年第5期33卷 1934-1945页

作者： Zhuang, Di Chang, J. Morris Li, Mingchen Univ S Florida Dept Elect Engn Tampa FL 33620 USA

Community detection is of great importance for online social network analysis. The volume, variety and velocity of data generated by today's online social networks are advancing the way researchers analyze those networks. For instance, real-world networks, such as Facebook, LinkedIn and Twitter, are inherently growing rapidly and expanding aggressively over time. However, most of the studies so far have been focusing on detecting communities on the static networks. It is computationally expensive to directly employ a well-studied static algorithm repeatedly on the network snapshots of the dynamic networks. We propose DynaMo, a novel modularity-based dynamic community detection algorithm, aiming to detect communities of dynamic networks as effective as repeatedly applying static algorithms but in a more efficient way. DynaMo is an adaptive and incremental algorithm, which is designed for incrementally maximizing the modularity gain while updating the community structure of dynamic networks. In the experimental evaluation, a comprehensive comparison has been made among DynaMo, Louvain (static) and 5 other dynamic algorithms. Extensive experiments have been conducted on 6 real-world networks and 10,000 synthetic networks. Our results show that DynaMo outperforms all the other 5 dynamic algorithms in terms of the effectiveness, and is 2 to 5 times (by average) faster than Louvain algorithm.

关键词： Heuristic algorithms Generators Facebook detection algorithms Clustering algorithms Machine learning algorithms Community detection dynamic network analysis modularity incremental approach

来源：评论

学校读者我要写书评

暂无评论

A Survey on Facial Wrinkles detection and Inpainting: Datasets, Methods, and Challenges

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2021年第4期5卷 505-519页

作者： Yap, Moi Hoon Batool, Nazre Ng, Choon-Ching Rogers, Mike Walker, Kevin Manchester Metropolitan Univ Dept Comp & Math Manchester M1 5GD England Scania CV AB Autonomous Transport Solut Predev & Res Sensors & Machine Learning Team SE-15187 Scania Sweden Panason R&D Ctr Singapore Singapore 469332 Singapore Image Metr Ltd Manchester M1 3HZ Lancs England

Automatic facial wrinkles detection and inpainting algorithms have gained attention of researchers in cosmetics, forensics and computer vision. The majority of current inpainting algorithms was implemented on the whole face, despite the fact that only imperfections need inpainting. In general, the definition of an imperfection is sign of ageing, spot, scar and freckles. This paper focuses on wrinkles as it is an obvious sign of ageing. We survey the computer vision techniques in facial wrinkles localisation from detection to inpainting. We present a comprehensive literature review on benchmark datasets, automated wrinkles detection algorithms and facial inpainting algorithms. Due to limited study on wrinkle inpainting, we inpaint the wrinkle regions using three state-of-the-art inpainting algorithms, namely flood-fill, Coherence Sensitivity Hashing and exemplar-based method. To assess the realism of inpainting results, we present the original and inpainted images to 40 participants, where they provide rating on the realism scale and age group of each image. The result shows that flood-fill method preserved the realism but there was no significant difference in age prediction. Finally, we conclude the paper by proposing some future directions to advance this field.

关键词： Faces Face recognition Skin detection algorithms Aging Forehead Estimation Facial wrinkles wrinkles detection wrinkles inpainting realism assessment

来源：评论

学校读者我要写书评

暂无评论

Convolutional Neural Network (CNN)-Based detection for Multi-Level-Cell NAND Flash Memory

引用

IEEE COMMUNICATIONS LETTERS 2021年第12期25卷 3883-3887页

作者： Shi, Zhifang Fang, Yi Bu, Yingcheng Han, Guojun Guangdong Univ Technol Sch Informat Engn Guangzhou 510006 Peoples R China Xidian Univ State Key Lab Integrated Serv Networks Xian 710126 Peoples R China

With the increase of program/erase (PE) cycles and retention time, it is difficult to predict the threshold-voltage distributions for detection in NAND flash memory. To accurately acquire the log-likelihood ratios (LLRs) without the knowledge of threshold-voltage distributions, a convolutional neural network (CNN)-based detection algorithm is proposed for the multi-level-cell (MLC) flash memory. The CNN-based detection algorithm employs the trained CNN to accurately calculate the LLRs for each threshold-voltage region. Furthermore, we develop a CNN-aided read-voltage design scheme to optimize the read voltages by maximizing the mutual information between the coded bits and their corresponding LLRs. Exploiting the proposed scheme, we first design three hard-decision read voltages, and then formulate more soft-decision read voltages to further improve the detection performance. Simulation results demonstrate that the CNN-based detection algorithm can achieve performance approaching that of the optimal detection algorithm.

关键词： Threshold voltage Ash detection algorithms Convolutional neural networks Training Voltage measurement Quantization (signal) NAND flash memory convolutional neural network read voltages detection

来源：评论

学校读者我要写书评

暂无评论

Design and Analysis of MIMO Systems Using Energy Detectors for Sub-THz Applications

引用

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS 2022年第6期21卷 3678-3690页

作者： Bicais, Simon Falempin, Alexis Dore, Jean-Baptiste Savin, Valentin Univ Grenoble Alpes CEA LETI F-38000 Grenoble France

The significant amount of unused spectrum in sub-TeraHertz frequencies is contemplated to realize high rate wireless communications for beyond 5G networks. Yet, the performance of radio-frequency sub-TeraHertz systems is severely degraded by strong oscillator phase noise. Therefore, we investigate in this paper the use of multiple-input multiple-output (MIMO) systems with energy detection receivers to achieve high rate communications robust to phase noise. First, the design of the receiver detection algorithm is addressed. Two detectors are proposed for the studied nonlinear MIMO channel, either derived from the maximum likelihood decision rule by using a Gaussian approximation, or based on the use of neural networks. Second, the communication performance is assessed through numerical simulations for uncoded and coded systems. We consider a realistic scenario modeling an indoor wireless link in D-band with directive antennas and strongly correlated line-of-sight channels. Our results demonstrate that spatial multiplexing with non-coherent sub-TeraHertz transceivers can be realized on strongly correlated line-of-sight channels using the proposed detection schemes. Thereby, we highlight that high-rate radio-frequency sub-TeraHertz systems can be implemented with low-complexity and low-power architectures using MIMO systems with energy detection receivers.

关键词： Radio frequency MIMO communication Detectors Wireless communication Neural networks Numerical models Phase noise Sub-Terahertz communications physical layer MIMO envelope detectors detection algorithms neural networks

来源：评论

学校读者我要写书评

暂无评论

Distributed Secure State Estimation for Cyber-Physical Systems Against Replay Attacks via Multisensor Method

引用

IEEE SYSTEMS JOURNAL 2022年第4期16卷 5720-5728页

作者： Su, Lei Ye, Dan Zhao, Xin-Gang Northeastern Univ Coll Informat Sci & Engn Shenyang 110819 Peoples R China Anhui Univ Technol Sch Elect & Informat Engn Maanshan 243032 Peoples R China Chinese Acad Sci Shenyang Inst Automat State Key Lab Robot Shenyang 110016 Peoples R China

This article investigates the distributed secure state estimation problem for cyber-physical systems under replay attacks. The main purpose is to design the distributed observers such that the estimation error system is stable. First, a detection algorithm is proposed to not only expose the behavior of the attacker but also identify which sensor is tampered. Then, the distributed observers are designed and the adverse effects are eliminated. Moreover, in view of the linear matrix inequality technique, a sufficient condition is presented, which ensures that the estimation error system is stable and achieves an $H_{\infty }$ property. Finally, a practical inverted pendulum example is employed to elaborate the effectiveness and superiority of the presented design and detection method.

关键词： Sensors Observers detection algorithms Watermarking Cyberattack Estimation error Time measurement detection mechanism distributed state estimation multisensor measurements replay attack

来源：评论

学校读者我要写书评

暂无评论

Fusing of a Continuous Output PDR Algorithm With an Ultrasonic Positioning System

引用

IEEE SENSORS JOURNAL 2022年第3期22卷 2464-2474页

作者： Thio, Vincent Aparicio, Joaquin Anonsen, Kjetil Bergh Bekkeng, Jan Kenneth Booij, Wilfred Univ Oslo Dept Phys N-0371 Oslo Norway Univ Oslo Dept Informat N-0373 Oslo Norway Norwegian Def Res Estab FFI N-2027 Kjeller Norway Univ Oslo Dept Technol Syst N-0371 Oslo Norway Forkbeard Technol AS N-0283 Oslo Norway

The availability of inertial navigation sensors in smartphones has facilitated the development of pedestrian dead reckoning (PDR) models on a large scale. These models often consist of a step detection algorithm combined with a heading estimation routine. Common approaches to step detection include searching for peaks/valleys in the acceleration signal, principle frequency estimation, and machine learning techniques. Since the sensors embedded in smart devices are prone to noise, the position error grows unbounded if unchecked and requires periodical corrections based on external measurements. In this work, we propose a novel step detection algorithm based on sine-wave approximation of the acceleration signal. This method detects step fractions as well as full steps, which allows for continuous and real-time updates. The step detection algorithm is combined with a heading estimation routine described in our previous work to obtain a stand-alone PDR model. To mitigate error accumulation, we fuse the proposed model with position and heading measurements provided by a commercial indoor positioning system based on ultrasound. We evaluate the performance of the PDR and fused model in an open office environment, by walking along a trajectory while carrying a smartphone in hand or in the pocket. The results demonstrate the feasibility of the sine-wave approximation approach to step detection, as well as the expected benefits of fusing PDR with the ultrasonic system.

关键词： Acoustics Accelerometers Sensors detection algorithms Position measurement Legged locomotion Estimation Indoor positioning pedestrian dead reckoning step detection ultrasonic positioning system

来源：评论

学校读者我要写书评

暂无评论

detection Algorithm for Cellular Synchronization Signals in Airborne Applications

引用

IEEE ACCESS 2021年 9卷 55555-55566页

作者： Stevens, Brian W. Younis, Mohamed F. Johns Hopkins Univ Appl Phys Lab Laurel MD 20723 USA Univ Maryland Baltimore Cty Dept Comp Sci & Elect Engn Baltimore MD 20723 USA

Cellular to air communication is critical for the booming aerial package delivery and transportation business. Detecting cellular signals in airborne applications is challenging because it requires receiving and processing waveforms that are subject to significantly more interference than those experienced in terrestrial settings. This paper highlights and tackles the complexity of 4G Long Term Evolution (LTE) signal synchronization in high altitude applications, e.g., cell access onboard an aircraft. Specifically, we design a novel cell detector that operates efficiently under high interference levels found in airborne applications, maintains a constant false alarm rate using an optimized threshold implementation for Zadoff Chu sequences, and monitors multiple towers with different time delays simultaneously. We validate our cell detector through simulation and experimentation. Lastly, the cell detector is used to estimate the interference in live waveforms taken from an aircraft at 2 to 2.5 km altitude and velocities of 200-400 km/h. Our cell detection model can be adapted to support 5G New Radio (NR) synchronization signals as NR deploys aerial support in the future. The threshold implementation to handle correlation spurs can be applied directly to other Zadoff Chu based signals such as random access signals found in both LTE and NR.

关键词： Synchronization Receivers Long Term Evolution Interference Time-frequency analysis OFDM Delays detection algorithms 4G long term evolution (LTE) 5G new radio (NR) aircraft unmanned aerial vehicles mobile communications correlation radio frequency signal detection cellular networks synchronization

来源：评论

学校读者我要写书评

暂无评论

Vulnerability Identification and Remediation of FDI Attacks in Islanded DC Microgrids Using Multiagent Reinforcement Learning

引用

IEEE TRANSACTIONS ON POWER ELECTRONICS 2022年第6期37卷 6359-6370页

作者： Abianeh, Ali Jafarian Wan, Yihao Ferdowsi, Farzad Mijatovic, Nenad Dragicevic, Tomislav Univ Louisiana Lafayette Dept Elect & Comp Engn Lafayette LA 70504 USA Tech Univ Denmark Dept Elect Engn DK-2800 Lyngby Denmark

This article proposes a novel approach to uncover deficiencies of the existing cyber-attack detection schemes and thereby to serve as a foundation for establishing more reliable cybersecure solutions, with particular application in dc microgrids. For this purpose, a multiagent deep reinforcement learning (RL)-based algorithm is proposed to automatically discover the vulnerable spots in the conventional index-based cyberattack detection schemes and automatically generate coordinated stealthy destabilizing false data injection (FDI) attacks on cyber-protected islanded dc microgrids. To enable a continuous action space for the trained RL agents and enhance the algorithm's precision and convergence rate, deep deterministic policy gradient is incorporated. Using this approach, susceptibility of a state-of-the-art detection scheme to several different coordinated FDI attacks on the distributed communication links is identified. The proposed algorithm is also enhanced with a sniffing feature to enable maintaining the stealthy attacks even under the sudden disconnection of any of the compromised links. To address the discovered deficiencies within the index-based detection scheme, a complementary multiagent RL detection algorithm using deep Q-network algorithm is integrated, which provides a more reliable overall identification performance. Taking into account the communication delays and load changes, the effectiveness of the proposed algorithm is verified by the experimental tests.

关键词： Microgrids Computer crime Reinforcement learning Decentralized control Voltage control Indexes detection algorithms Cybersecurity dc microgrid distributed control false data injection (FDI) reinforcement learning (RL)

来源：评论

学校读者我要写书评

暂无评论

A Message-Based Malicious detection Scheme of Public DNS Services 23

A Message-Based Malicious Detection Scheme of Public DNS Ser...

引用

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on Data Science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Li, Chaoqun Dai, Liang Xu, Zhen Ding, Ying Han, Yanni Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781665494571

The open resolver, which is an important component in the DNS system, can provide resolution services for the public. It is widely accepted that the public resolution service they use is safe. But in fact, this trust in resolution services is extremely blind and unreliable. Malicious open resolvers on the Internet tamper with user requests and direct users' normal DNS requests to the wrong destination. Therefore, it is necessary to discover these malicious attacks for Internet users as soon as possible. The traditional response strategy is to use a blacklist, which is simple to implement while cannot be exhaustive. Simultaneously, it cannot be applied to the dynamic situation in which the address of the malicious open resolver changes under actual network conditions. In this paper, by analyzing the correlations between DNS messages and public resolution, we propose features for malicious behavior detection. Furthermore, we design a malicious open resolver detection method based on random forest. To the best of our knowledge, our work conducts the first study to counter malicious open resolvers in this field. Experiment results demonstrate the ability to detect malicious open resolvers qualitatively. By using the features based on DNS messages, our detection method has a true positive rate of 99.53% and a low false positive rate of 0.87%, which verifies the effectiveness of our method. © 2021 IEEE.

关键词： Correlation Forestry Feature extraction Behavioral sciences Internet Blocklists detection algorithms

来源：评论

学校读者我要写书评

暂无评论

To Develop Multi-Object detection and Recognition Using Improved GP-FRCNN Method 8

To Develop Multi-Object Detection and Recognition Using Impr...

引用

8th International Conference on Smart Structures and Systems, ICSSS 2022

作者： Loganathan, Ganesh Babu Fatah, Tooraj Hassan Yasin, Elham Tahsin Hamadamen, Nawroz Ibrahim Tishk International University Faculty of Engineering Department of Mechatronics Kurdistan Region Erbil Iraq Noble Private Institute Information Technology Kurdistan Region Erbil Iraq University of Salahaddin Institute of Higher Education & Scientific Research Software Engineering Department Kurdistan Region Erbil Iraq

ISBN: (纸本)9781665497619

The Region-based Convolutional Neural Network (R-CNN), which is more rapid, has been recognized as the best detection algorithm. With our design basis for Faster RCNN (FRCNN), the designers reclassify the proposed object recognition changes with just a rough spatial estimate of the incident. The researchers solve problems such as delicate to complex environments and its principles are derived reductions each time it is -actually confronted with noisier initiatives. Through which designers categorize proposed changes to object recognition using a rough geometric estimate of the scene. This simple expansion requires complex changes of scale, which the authors describe in their entirely throughout this study. Moreover, Geometric Proposals with FRCNN (GP-FRCNN) performed relatively well on small and large things, which has always been a problem for computer vision algorithms. Any use of GP-FRCNN for monitoring film analysis is simple and requires no specific geometric description. We tested the algorithm on the UA-DETRAC database, finding that GP-FRCNN beat the FRCNN by 21%. © 2022 IEEE.

关键词： Training Image recognition Databases Object recognition Proposals Intelligent structures detection algorithms

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：