检索结果-内蒙古大学图书馆

2023 IEEE/CVF Conference on computer Vision and Pattern Recognition, CVPR 2023

作者： Zhu, Zijian Zhang, Yichi Chen, Hai Dong, Yinpeng Zhao, Shu Ding, Wenbo Zhong, Jiachen Zheng, Shibao Institute of Image Communication and Network Engineering Shanghai Jiao Tong University China Institute for Ai Tsinghua University BNRist Center Thbi Lab Dept. of Comp. Sci. and Tech. China School of Computer Science and Technology Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Information Materials and Intelligent Sensing Laboratory of Anhui Province China Saic Motor Ai Lab Zhongguancun Laboratory China

ISBN: (纸本)9798350301298

3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations;2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features;3) Camera-LiDARfusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models. © 2023 IEEE.

关键词： Autonomous driving

来源：评论

学校读者我要写书评

暂无评论

A high-resolution DOA estimation method with a family of nonconvex penalties

arXiv

引用

arXiv 2017年

作者： Wu, Xiaohuan Zhu, Wei-Ping Yan, Jun Key Lab of Broadband Wireless Communication and Sensor Network Technology Nanjing University of Posts and Telecommunications Nanjing210003 China Department of Electrical and Computer Engineering Concordia University Montreal Canada School of Communication and Information Engineering Nanjing University of Posts and Telecommunications Nanjing China

The low-rank matrix reconstruction (LRMR) approach is widely used in direction-of-arrival (DOA) estimation. As the rank norm penalty in an LRMR is NP-hard to compute, the nuclear norm (or the trace norm for a positive semidefinite (PSD) matrix) has been often employed as a convex relaxation of the rank norm. However, solving a nuclear norm convex problem may lead to a suboptimal solution of the original rank norm problem. In this paper, we propose to apply a family of nonconvex penalties on the singular values of the covariance matrix as the sparsity metrics to approximate the rank norm. In particular, we formulate a nonconvex minimization problem and solve it by using a locally convergent iterative reweighted strategy in order to enhance the sparsity and resolution. The problem in each iteration is convex and hence can be solved by using the optimization toolbox. Convergence analysis shows that the new method is able to obtain a suboptimal solution. The connection between the proposed method and the sparse signal reconstruction (SSR) is explored showing that our method can be regarded as a sparsity-based method with the number of sampling grids approaching infinity. Two feasible implementation algorithms that are based on solving a duality problem and deducing a closed-form solution of the simplified problem are also provided for the convex problem at each iteration to expedite the convergence. Extensive simulation studies are conducted to show the superiority of the proposed methods. Copyright © 2017, The Authors. All rights reserved.

关键词： Covariance matrix

来源：评论

学校读者我要写书评

暂无评论

Finformer: A Static-dynamic Spatiotemporal Framework for Stock Trend Prediction

Finformer: A Static-dynamic Spatiotemporal Framework for Sto...

引用

IEEE International Conference on Big Data

作者： Yi Zu Jiacong Mi Lingning Song Shan Lu Jieyue He School of Computer Science and Engineering Key Lab of Computer Network and Information Integration MOE Southeast University Nanjing China School of Software Engineering Southeast University Nanjing China Nanjing Fenghuo Tiandi Communication Technology Co. Ltd Nanjing China

The core of quantitative investment lies in predicting future trends in stock prices. The future trend of a stock is closely related to the industry it belongs to and its relationship with other stocks. Although some research has focused on stock trend prediction in recent years, most studies have only considered the stock’s own time series feature, neglecting the spatial features between stocks. Some research has incorporated spatial information, but typically only considered predefined static relationships. At the same time, capturing dynamic spatial information in the market has been a long-standing challenge. Thus, we propose a spatio-temporal model, Finformer, in order to go beyond traditional time series models. We designed a sparse static-dynamic transformer to capture dynamic market spatial information as it changes over time and combined predefined relationships to extract highly correlated spatial features in the stock market. To effectively integrate spatial and temporal features, we introduced an adaptive spatio-temporal fusion module that dynamically fuses spatio-temporal features based on market conditions at different periods. Experiments on two real-world stock market datasets show that our proposed model outperforms the state-of-the-art baselines in the signal-based and portfolio-based metrics, which are widely concerned in the financial field. Ablation study and hyper-parameter study further reveal the effectiveness of each module in the model and the impact of hyper-parameters. The code will be made publicly available. 1

关键词：

来源：评论

学校读者我要写书评

暂无评论

Differential space-time block coding based cooperative spectrum sensing over fading environments in cognitive radio sensor networks

引用

Journal of Information and Computational Science 2012年第15期9卷 4599-4606页

作者： Chen, Zhi Yue, Wenjing College of Computer Nanjing University of Posts and Telecommunications Nanjing 210003 China Key Laboratory for Novel Software Technology Nanjing University Nanjing 210093 China Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks Nanjing 210003 China Key Lab. of Broadband Wireless Communication and Sensor Network Technology Nanjing University of Posts and Telecommunications Ministry of Education Nanjing 210003 China

Incorporating cognitive radio capability in recource-constrainted multihop wireless sensor networks may help realize the deployment of multiple overlaid sensor networks, and eliminate collision and excessive contention delay incurred by dense node deployment. One of the main requirements of cognitive radio sensor networks is the ability to detect the presence of the primary user with fast speed and high accuracy in wireless environments. Cooperative spectrum sensing emerges as a promising candidate to meet these requirements. In practice, however, since the local sensing observations are forwarded to a sink through fading channels, the sensing performance of cooperative spectrum sensing is deteriorated due to the reporting errors introduced by the fading channels. To deal with this limitation, we propose a differential space-time block coding based cooperative spectrum sensing scheme to improve the sensing performance by decreasing the probability of reporting error. The sensing performance is also derived. Numerical results verify that better sensing performance can be achieved in proposed scheme compared to the conventional cooperative spectrum sensing. © 2012 by Binary Information Press.

关键词： Cognitive radio

来源：评论

学校读者我要写书评

暂无评论

Bacterial foraging optimization algorithm based routing strategy for wireless sensor networks

引用

International Review on computers and Software 2012年第6期7卷 2826-2830页

作者： Chen, Zhi Li, Shuai Yue, Wenjing Hu, Luoquan Sun, Wanxin College of Computer Nanjing University of Posts and Telecommunications Nanjing 210003 China State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210093 China Key Lab of Broadband Wireless Communication and Sensor Network Technology Ministry of Education Nanjing 210003 China Suzhou Entry-Exit Inspection and Quarantine Bureau Suzhou 215104 China

In order to improve energy efficiency, the bacterial foraging optimization algorithm (OBFOA) is applied into the routing strategy for wireless sensor networks. The proposed strategy makes proper improvements on bacteria's foraging operation to improve the performance of the algorithm, such as the adaptive step strategy in the swimming operation and the comprehensive selection mechanism in the tumbling operation. OBFOA builds optimized data transfer paths in wireless sensor networks through the processes of chemotaxis, global quorum sensing mechanisms, reproduction and elimination-dispersal. Simulation results show that the proposed strategy can balance the energy consumption of nodes effectively and prolong the overall network lifetime. © 2012 Praise Worthy Prize S.r.l. - All rights reserved.

关键词： Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Wavelet-analysis based file-sharing P2P traffic prediction algorithm

International Journal of Digital Content Technology and its ...

引用

International Journal of Digital Content technology and its Applications 2011年第6期5卷 46-54页

作者： Wu, Min Wang, Ruchuan Han, Zhijie College of Computer Nanjing University of Posts and Telecommunications Nanjing Jiangsu 210003 China Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks Nanjing Jiangsu 210003 China Key Lab of Broadband Wireless Communication and Sensor Network Technology Nanjing University of Posts and Telecommunications Ministry of Education Jiangsu Province Nanjing Jiangsu 210003 China

The increasing P2P network traffic on the Internet has leaded to the problem of network congestion. In the consequence of the diversification of the P2P business and protocol, research on the management of P2P traffic has many problems to resolve. Prediction of the P2P traffic is the kernel problem in the P2P traffic management. Based on the existed P2P traffic characters, this paper structures a P2P traffic model, gives a traffic prediction algorithm bases on wavelet-analysis, and proves the accuracy of the algorithm. Simulation experiment figures that the algorithm has a high prediction precision and a superior real-time performance.

关键词： Wavelet analysis

来源：评论

学校读者我要写书评

暂无评论

Risk prediction model based on improved AdaBoost method for cloud users

引用

Open Cybernetics and Systemics Journal 2015年第1期9卷 44-49页

作者： Zhang, Lin Rao, Kaili Wang, Ruchuan Jia, Yishang College of Computer Nanjing University of Posts and Telecommunications Nanjing210003 China Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks NanjingJiangsu Province210003 China Key Lab of Broadband Wireless Communication and Sensor Network Technology Nangjing University of Posts and Telecommunications Ministry of Education Jiangsu Province Nangjing210003 China

Considering the problem how to protect the cloud services from being destroyed by cloud users, the riskprediction model based on improved AdaBoost method is proposed. The risk prediction is regarded as two-class classification problem, and the risk of new cloud users could be predicted by the attributes of historical cloud users. In order to improve the result of predicted, AdaBoost method is adopted in this paper. The error rate of the last training is used to adjust the sample distribution of the next training, which can make the next training have stronger ability of identification for the error-classified samples. At the same time the weight of each weak classifier is set. After all, the strong classifier is generated by combined the weak classifiers through voting, which can improve the overall result of classification. Considering the wrongly-predicted cost, AdaBoost method is improved. The method of cost-sensitive is added into the model in order to minimal the misclassified-cost. Experiments show that the cost-sensitive AdaBoost method has better classification result than the traditional ones and it can predict the risk of the new cloud user effectively and protect the security of the cloud services. © Zhang et al.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

An -norm Based Method for Off-grid DOA Estimation

引用

Circuits, Systems, and Signal Processing 2018年第2期38卷 904-917页

作者： Zhang, Zeyun Wu, Xiaohuan Li, Chunguo Zhu, Wei-Ping The Key Lab of Broadband Wireless Communication and Sensor Network Technology Nanjing University of Posts and Telecommunications Nanjing People’s Republic of China School of Information Science and Engineering Southeast University Nanjing People’s Republic of China Department of Electrical and Computer Engineering Concordia University Montreal Canada

The sparse signal recovery-based direction of arrival (DOA) estimation has received a great deal of attention over the past decade. From the sparse representation point of view, $$\ell _0$$ -norm is the best choice to evaluate the sparsity of a vector. However, solving an $$\ell _0$$ -norm minimization problem is non-deterministic polynomial hard (NP-hard). Thus, The common idea for many sparse DOA estimation methods is to use the $$\ell _1$$ -norm as the sparsity metric. However, its sparse solution may not coincide with the solution resulting from the $$\ell _0$$ -norm thus deteriorating the DOA estimation performance. In this paper, we propose a new sparse method based on $$\ell _p$$ ( $$0<1$$ ) regularization for DOA estimation to achieve a sparser solution than $$\ell _1$$ regularization. In particular, we use the Taylor expansion to convert the $$\ell _p$$ -norm minimization problem to a weighted $$\ell _1$$ -norm problem. Then, a two-step iterative method is employed to achieve the DOA estimate. The $$\ell _p$$ ( $$0<1$$ ) regularization is able to improve the angle resolution, leading to an improved performance in low SNR and correlated signal scenarios. Numerical results show that our proposed method has better estimation performance than many other methods do.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Study on Object Transaction Based on networked RFID Systems

Study on Object Transaction Based on Networked RFID Systems

引用

2014 IEEE 5th International Conference on Software Engineering and Service Science

作者： Xuan Shen Chenqi Bao Jianmin Bao Huan Wang Department of Computer and Software Nanjing University of Posts and Telecommunications Key Lab of Broadband Wireless Communication and Sensor Network Technology Ministry of EducationCollege of IOT Nanjing University of Posts and Telecommunications Department of IOT Nanjing University of Posts and Telecommunications

ISBN: (纸本)9781479932801

Radio Frequency IDentification(RFID) is a novel technology of automated *** use of RFID,however,raises great concerns due to the serious security *** this paper,we at first propose a Secure Transaction Frame(STF) and a flow of object transaction based on networked RFID ***,it is worth mentioning that the authentication technology in Secure Transaction Frame(STF) is able to verify the ownership of electric tags in order to deal with common attacks in the networked RFID ***,this paper presents an encryption scheme to protect the information stored in electric tags from being illegally *** collecting relevant statistics and analyzing cases based on networked RFID systems,this paper also researches the mass requests for the information stored in the electric tags by the reader and eventually,gives a cache mechanism to solve that problem.

关键词： RFID object transaction secure transaction flow(STF) mass requests cache technology

来源：评论

学校读者我要写书评

暂无评论

An Improved DV-Hop Positioning Algorithm for Wireless Sensor network

An Improved DV-Hop Positioning Algorithm for Wireless Sensor...

引用

2015 IEEE International Conference on Progress in Informatics and Computing(PIC 2015)

作者： Jiuhu Zheng Huanyan Qian Lie Wang School of Computer Science and Engineering Nanjing University of Science and Technology School of Information Technology Jiangsu Institute of Commerce Key Lab of Broadband Wireless Communication and Sensor Network Technology ( Nanjing University of Posts and Telecommunications) Ministry of Education National Mobile Communications esearch Laboratory Southeast University

In the view of the DV-Hop positioning algorithm which relies on the distance vector and beacon nodes, several aspects are improved: the average hop distance and hop count are corrected;a method to detect the abnormal distribution of beacon nodes is used to avoid the influence of the abnormal distribution on the positioning performance. The proposed algorithm is simple and effective. Simulation results show that compared with the traditional DV-Hop algorithm, the performance of the proposed algorithm is greatly *** performance is even even superior to that of the algorithm mentioned in literature [10]

关键词： Wireless sensor network DV-Hop algorithm node positioning positioning accuracy

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：