检索结果-南通市图书馆

Small-signal stability and robustness analysis for microgrids under time-constrained DoS attacks and a mitigation adaptive secondary control method

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2022年第6期65卷 107-124页

作者： Qiuye SUN Bingyu WANG Xiaomeng FENG Shiyan HU College of Information science and Engineering Northeastern University School of Electronics and Computer Science University of Southampton

With the close integration of cyber and power systems, the consensus-based secondary frequency control in a microgrid is increasingly vulnerable to communication failures such as transmission delays and denial-of-service(DoS) attacks, which can affect the efficiency of frequency recovery in the secondary frequency control. Leveraging the small-signal model, this paper develops a novel cyber-physical system model to analyze the cross-layer effect of DoS attacks on microgrids. In this way, the cross-layer impact on the microgrid from the cyber system to the physical system can be convincingly analyzed. Based on the root approximation method, the tolerant saving time is designed for the microgrid as the index to evaluate the tolerance margin of the time-constrained DoS attack, and then the relationship between the margin and secondary control coefficients is found. A mitigation adaptive secondary control technique is proposed so that the attacked microgrid can dynamically tune the secondary control gain according to the saving time and tolerant saving time(TST). The simulation results show that although the microgrid with high secondary control gain has good dynamic robustness, its TST is low. In addition, the proposed adaptive secondary control system is significantly better than the traditional control system in terms of the stability performance of the microgrid under a DoS attack.

关键词： microgrid hierarchical control architecture cyber-physical system security denial-of-service attack adaptive secondary frequency control

LSTM-based argument recommendation for non-API methods

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2020年第9期63卷 5-26页

作者： Guangjie LI Hui LIU Ge LI Sijie SHEN Hanlin TANG school of computer science and Technology Beijing Institute of Technology school of electronics Engineering and computer science Peking University

Automatic code completion is one of the most useful features provided by advanced IDEs. Argument recommendation, as a special kind of code completion, is widely used as well. While existing approaches focus on argument recommendation for popular APIs, a large number of non-API invocations are requesting for accurate argument recommendation as well. To this end, we propose an LSTM-based approach to recommending non-API arguments instantly when method calls are typed in. With data collected from a large corpus of open-source applications, we train an LSTM neural network to recommend actual arguments based on identifiers of the invoked method, the corresponding formal parameter, and a list of syntactically correct candidate arguments. To feed these identifiers into the LSTM neural network, we convert them into fixed-length vectors by Paragraph Vector, an unsupervised neural network based learning algorithm. With the resulting LSTM neural network trained on sample applications, for a given call site we can predict which of the candidate arguments is more likely to be the correct one. We evaluate the proposed approach with tenfold validation on 85 open-source C applications. Results suggest that the proposed approach outperforms the state-of-the-art approaches in recommending non-API arguments. It improves the precision significantly from 71.46% to 83.37%.

关键词： argument recommendation LSTM deep learning non-API

Deep learning framework for multi‐round service bundle recommendation in iterative mashup development

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2023年第3期8卷 914-930页

作者： Yutao Ma Xiao Geng Jian Wang Keqing He Dionysis Athanasopoulos school of computer science Wuhan UniversityWuhanChina school of computer science Central China Normal UniversityWuhanChina school of electronics Electrical Engineering and Computer ScienceQueen's University BelfastBelfastUK

Recent years have witnessed the rapid development of service‐oriented computing technologies.The boom of Web services increases software developers'selection burden in developing new service‐based systems such as mashups.Timely recommending appropriate component services for developers to build new mashups has become a fundamental problem in service‐oriented software engineering.Existing service recom-mendation approaches are mainly designed for mashup development in the single‐round scenario.It is hard for them to effectively update recommendation results according to developers'requirements and behaviours(e.***.instant service selection).To address this issue,the authors propose a service bundle recommendation framework based on deep learning,DLISR,which aims to capture the interactions among the target mashup to build,selected(component)services,and the following service to recommend.Moreover,an attention mechanism is employed in DLISR to weigh selected services when rec-ommending a candidate service.The authors also design two separate models for learning interactions from the perspectives of content and invocation history,respectively,and a hybrid model called HISR.Experiments on a real‐world dataset indicate that HISR can outperform several state‐of‐the‐art service recommendation methods to develop new mashups iteratively.

关键词： attention deep learning mashup development recommender systems service bundle

An Improved Method of Ambiguity Resolution in GNSS Positioning

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of electronics 2019年第1期28卷 215-222页

作者： GU Xiaoying ZHU Bocheng school of electronics Engineering and computer science Peking University

High precision navigation with positioning accuracy of decimeter or even centimeter is becoming more and more important in many fields. Accurate carrier phase observations are the prerequisites for this requirement. Because they have much lower measurement noises than the code observations which are primarily used nowadays. However, the existence of integer ambiguity when counting cycles of carrier phases prohibits the straightforward application of the measurements. Once the integer ambiguity has been resolved, the application of carrier phase measurement is almost equivalent to pseudorange however with much higher precision. Therefore ambiguity resolution is key for fast and high precision positioning. The success rate of ambiguity resolution is affected by the ionospheric delay and observation noise. In previous methods, ionospheric errors are simply ignored. To resolve ambiguity reliably and quickly, we propose a modified method for precise point positioning.We analyze the characteristics of the dual-frequency combinations of original observations and the ones with longer wavelengths and lower noises are preferable. The wide-lane and sub-wide-lane combinations with higher success rates are chosen for ambiguity resolution. Extra pseudoranges are included to eliminate the ionospheric delay which hampers ambiguity resolution seriously. After the combined ambiguities have been resolved, the original ambiguities of each frequency can be calculated from the linearly independent equations. Based on real Global positioning system(GPS) navigation data, performances of the modified method are tested and compared with that of the traditional method. The results show that the modified method is less affected by ionospheric delay and can obtain more accurate ambiguity resolution.

关键词： Global navigation satellite system Integer ambiguity resolution Precise point positioning High precision positioning

A Simultaneous Cartoon-Texture Image Segmentation and Image Decomposition Method

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of electronics 2020年第5期29卷 906-915页

作者： LI Yafeng ZHAO Qijun ZHANG Wenbo FAN Pan ZHANG Renrui SUN Jieqi LI Jing school of computer science and Technology Baoji University of Arts and Science College of computer science Sichuan University school of Information science and Technology Tibet University school of electronics Engineering and computer science Peking University

Image segmentation and image decomposition are fundamental problems in image processing.Image decomposition methods for separating images into cartoon and texture components can effectively serve different image processing tasks because different components can be respectively treated in more effective way. However, image decomposition methods are currently simply taken as an independent preprocessing step, and particularly in image segmentation different effects of cartoon and texture components have not been considered.This paper presents a novel simultaneous cartoon-texture image segmentation and image decomposition method to boost the performance of both segmentation and decomposition. We design a fast alternating optimization algorithm to solve the proposed model. Experimental results demonstrate the outstanding performance of the proposed method on both image segmentation and image decomposition.

关键词： Image segmentation Image decomposition Variational model Dictionary learning

Abusive adversarial agents and attack strategies in cyber-physical systems

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2023年第1期8卷 149-165页

作者： Viraj Singh Gaur Vishal Sharma John McAllister school of electronics Electrical Engineering and Computer ScienceQueen's University BelfastBelfastUK

The exponential increase in IoT device usage has spawned numerous cyberspace innovations.IoT devices,sensors,and actuators bridge the gap between physical processes and the cyber network in a cyber-physical system(CPS).Cyber-physical system is a complex system from a security perspective due to the heterogeneous nature of its components and the fact that IoT devices can serve as an entry point for cyberattacks.Most adversaries design their attack strategies on systems to gain an advantage at a relatively lower cost,whereas abusive adversaries initiate an attack to inflict maximum damage without regard to cost or reward.In this paper,a sensor spoofing attack is modelled as a malicious adversary attempting to cause system failure by interfering with the feedback control mechanism.It is accomplished by feeding spoofed sensor values to the controller and issuing erroneous commands to the actuator.Experiments on a Simulink-simulated linear CPS support the proof of concept for the proposed abusive ideology,demonstrating three attack strategies.The impact of the evaluations stresses the importance of testing the CPS security against adversaries with abusive settings for preventing cyber-vandalism.Finally,the research concludes by highlighting the limitations of the proposed work,followed by recommendations for the future.

关键词： intelligent systems security security evaluation

Robust sparse representation based face recognition in an adaptive weighted spatial pyramid structure

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2018年第1期61卷 86-98页

作者： Xiao MA Fandong ZHANG Yuelong LI Jufu FENG school of electronics Engineering and computer science Peking University school of computer science and Software Engineering Tianjin Polytechnic University Department of computer science University of York

The sparse representation based classification methods has achieved significant performance in recent years. To fully exploit both the holistic and locality information of face samples, a series of sparse representation based methods in spatial pyramid structure have been proposed. However, there are still some limitations for these sparse representation methods in spatial pyramid structure. Firstly, all the spatial patches in these methods are directly aggregated with same weights, ignoring the differences of patches’ reliability. Secondly, all these methods are not quite robust to poses, expression and misalignment variations, especially in under-sampled cases. In this paper, a novel method named robust sparse representation based classification in an adaptive weighted spatial pyramid structure(RSRC-ASP) is proposed. RSRC-ASP builds a spatial pyramid structure for sparse representation based classification with a self-adaptive weighting strategy for residuals’ aggregation. In addition, three strategies, local-neighbourhood representation, local intra-class Bayesian residual criterion, and local auxiliary dictionary, are exploited to enhance the robustness of RSRC-ASP. Experiments on various data sets show that RSRC-ASP outperforms the classical sparse representation based classification methods especially for under-sampled face recognition problems.

关键词： face recognition sparse representation self-adaptive weighted aggregating spatial pyramid structure local robust strategies

Stabilized DFB laser system with large tuning range

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Optics Letters 2018年第3期16卷 35-39页

作者：于佳晨王平俊于齐张胤熊炜陈徐宗 Institute of Quantum electronics School of Electronics Engineering and Computer Science Peking University

We propose, design, and realize a compact stabilized laser system that can be tuned within 24 GHz automatically. This laser system consists of two distributed feedback （DFB） lasers, one of which is reference and locked to the D2 line of Rb87, the other laser is a slave that is locked to the reference laser via a loop servo. We measured the frequency of the beating signal of the two lasers and generated an error signal, which controlled the frequency of the slave laser to close the loop. We compressed the fluctuation of the beating signal’s frequency to less than 1 MHz. Furthermore, the system can also automatically determine and control whether the slave is red detuned or blue detuned to the reference. The dimensions of our laser system are about 15 cm×20 cm×10 cm. This kind of laser system can be applied in many important applications, such as atomic interferometer and cold atomic clock.

关键词： DFB Stabilized DFB laser system with large tuning range

维普期刊数据库

Collaborative representation Bayesian face recognition

在线全文

学校读者我要写书评

暂无评论

science China(Information sciences) 2017年第4期60卷 232-234页

作者： Jufu FENG Xiao MA Wenjing ZHUANG school of electronics Engineering and computer science Peking University

In recent years,a series of subspace methods,named the collaborative representation based methods,have aroused researchers’interests.Inspired by the idea of sparse coding,Wright et al.[1]proposed the sparse representation based classification(SRC)method.SRC encodes a query sample as a linear combination of all subjects’training samples with sparsity constraint and then classifies it by evaluating which class has the minimal

关键词： SRC CRC