检索结果-南通市图书馆

A robust and efficient algorithm for Chinese historical document analysis and recognition

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

National Science Review 2023年第6期10卷 9-12页

作者： Chongyu Liu Cheng Jian Jiarong Huang Wentao Yang Yongxin Shi Qing Jiang Lianwen Jin School of Electronic and Information Engineering South China University of Technology SCUT-Zhuhai Institute of Modern Industrial Innovation

PROBLEM Historical documents,which contain valuable information on historical knowledge and literary arts,are important mediums for the inheritance of human ***,the preservation of historical documents is considered to be an urgent research *** efficient way is to develop a document digitization system,which requires accurate historical document recognition and *** the rapid development of optical character recognition techniques,historical document understanding has made great progress [1,2] and several benchmarkshave been established [3-5].

关键词：

An E-band CMOS frequency quadrupler with 1.7-dBm output power and 45-dB fundamental suppression

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Semiconductors 2022年第9期43卷 46-52页

作者： Xiaofei Liao Dixian Zhao Xiaohu You National Mobile Communication Research Laboratory Southeast UniversityNanjing 211189China Purple Mountain Laboratories Nanjing 211111China

This paper presents an E-band frequency quadrupler in 40-nm CMOS *** circuit employs two push-push frequency doublers and two single-stage neutralized *** pseudo-differential class-B biased cascode topo-logy is adopted for the frequency doubler,which improves the reverse isolation and the conversion *** tech-nique is applied to increase the stability and the power gain of the amplifiers *** stacked transformers are used for single-ended-to-differential transformation as well as output bandpass *** output bandpass filter enhances the 4th-harmonic output power,while rejecting the undesired harmonics,especially the 2nd *** core chip is 0.23 mm^(2)in size and consumes 34 *** measured 4th harmonic achieves a maximum output power of 1.7 dBm with a peak conversion gain of 3.4 dB at 76 *** fundamental and 2nd-harmonic suppressions of over 45 and 20 dB are achieved for the spectrum from 74 to 82 GHz,respectively.

关键词： capacitor neutralization CMOS E-band frequency doubler frequency quadrupler push-push

Towards imbalanced motion:part-decoupling network for video portrait segmentation

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第7期67卷 197-210页

作者： Tianshu YU Changqun XIA Jia LI State key Laboratory of Virtual Reality Technology and Systems School of Computer Science and EngineeringBeihang University Peng Cheng Laboratory

Video portrait segmentation(VPS), aiming at segmenting prominent foreground portraits from video frames, has received much attention in recent years. However, the simplicity of existing VPS datasets leads to a limitation on extensive research of the task. In this work, we propose a new intricate large-scale multi-scene video portrait segmentation dataset MVPS consisting of 101 video clips in 7 scenario categories,in which 10843 sampled frames are finely annotated at the pixel level. The dataset has diverse scenes and complicated background environments, which is the most complex dataset in VPS to our best *** the observation of a large number of videos with portraits during dataset construction, we find that due to the joint structure of the human body, the motion of portraits is part-associated, which leads to the different parts being relatively independent in motion. That is, the motion of different parts of the portraits is imbalanced. Towards this imbalance, an intuitive and reasonable idea is that different motion states in portraits can be better exploited by decoupling the portraits into parts. To achieve this, we propose a part-decoupling network(PDNet) for VPS. Specifically, an inter-frame part-discriminated attention(IPDA)module is proposed which unsupervisedly segments portrait into parts and utilizes different attentiveness on discriminative features specified to each different part. In this way, appropriate attention can be imposed on portrait parts with imbalanced motion to extract part-discriminated correlations, so that the portraits can be segmented more accurately. Experimental results demonstrate that our method achieves leading performance with the comparison to state-of-the-art methods.

关键词： video portrait segmentation imbalanced motion unsupervised part decoupling motion correlation inter-frame attention

An Energy-Efficient UAV Deployment Scheme for Emergency Communications in Air-Ground Networks with Joint Trajectory and Power Optimization

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

China Communications 2022年第7期19卷 67-78页

作者： Shuo Zhang Shuo Shi Weizhi Wang Zhenyu Xu Xuemai Gu School of Electronic and Information Engineering Harbin Institute of TechnologyHarbin 150001China Peng Cheng Laboratory Shenzhen 518055China Department of Intelligent Engineering Huizhou Engineering Vocational CollegeHuizhou 516023China

The space-air-ground integrated network(SAGIN)has gained widespread attention from academia and industry in recent *** is widely applied in many practical fields such as global observation and mapping,intelligent transportation systems,and military *** an information carrier of air platforms,the deployment strategy of unmanned aerial vehicles(UAVs)is essential for communication systems’*** this paper,we discuss a UAV broadcast coverage strategy that can maximize energy efficiency(EE)under terrestrial users’*** to the non-convexity of this issue,conventional approaches often solve with heuristics algorithms or alternate *** this end,we propose an iterative algorithm by optimizing trajectory and power allocation ***,we discrete the UAV trajectory into several stop points and propose a user grouping strategy based on the traveling salesman problem(TSP)to acquire the number of stop points and the optimization ***,we use the Dinkelbach method to dispose of the fractional form and transform the original problem into an iteratively solvable convex optimization problem by variable substitution and Taylor *** results validate our proposed solution and outperform the benchmark schemes in EE and mission completion time.

关键词： SAGIN UAV energy efficiency(EE)maximization trajectory optimization power allocation

Extrapolation over temporal knowledge graph via hyperbolic embedding

在线全文

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2023年第2期8卷 418-429页

作者： Yan Jia Mengqi Lin Ye Wang Jianming Li Kai Chen Joanna Siebert Geordie Z.Zhang Qing Liao Pengcheng Laboratory ShenzhenChina Harbin Institute of Technology(Shenzhen) ShenzhenChina National University of Defense Technology ChangshaChina The University of Melbourne MelbourneAustralia

Predicting potential facts in the future,Temporal Knowledge Graph(TKG)extrapolation remains challenging because of the deep dependence between the temporal association and semantic patterns of ***,facts(events)that happened at different timestamps have different influences on future events,which can be attributed to a hierarchy among not only facts but also relevant ***,it is crucial to pay more attention to important entities and events when forecasting the ***,most existing methods focus on reasoning over temporally evolving facts or mining evolutional patterns from known facts,which may be affected by the diversity and variability of the evolution,and they might fail to attach importance to facts that *** geometry was proved to be effective in capturing hierarchical patterns among data,which is considered to be a solution for modelling hierarchical relations among *** this end,we propose ReTIN,a novel model integrating real-time influence of historical facts for TKG reasoning based on hyperbolic geometry,which provides low-dimensional embeddings to capture latent hierarchical structures and other rich semantic patterns of the existing *** both real-time and global features of TKG boosts the adaptation of ReTIN to the ever-changing dynamics and inherent *** experiments on benchmarks demonstrate the superiority of ReTIN over various *** ablation study further supports the value of exploiting temporal information.

关键词： extrapolation hyperbolic embedding temporal knowledge graph

Ultra-wideband fiber-THz-fiber seamless integration communication system toward 6G: architecture, key techniques, and testbed implementation

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) 2023年第1期66卷 296-313页

作者： Min ZHU Jiao ZHANG Bingchang HUA Mingzheng LEI Yuancheng CAI Liang TIAN Dongming WANG Wei XU Chuan ZHANG Yongming HUANG Jianjun YU Xiaohu YOU National Mobile Communications Research Laboratory Southeast University Purple Mountain Laboratories School of Information Science and Technology Fudan University Peng Cheng Laboratory

Terahertz(THz) communication is widely regarded as the key component of future 6G mobile communication systems. Through comparative analysis of some of the main existing technical routes of THz up-conversion and down-conversion in THz wireless communication systems, a novel ultra-wideband(UWB)fiber-THz-fiber seamlessly converged real-time architecture, which utilizes the commercially mature digital coherent optical module to realize ultrahigh-capacity THz real-time wireless communication, is proposed in this study.(1) The proposed architecture employs the dual-polarization photonic up-conversion technique for THz generation and hybrid optoelectronic down-conversion technique for THz reception to facilitate the seamless integration between optical fiber and THz communications.(2) Because of the limited bandwidth of optoelectronic devices, multidimensional modulation techniques are adopted for UWB THz signals to improve spectral efficiency and transmission capacity.(3) An intelligent nonlinear joint compensation technique based on the deep neural network, which can effectively improve the signal-to-noise ratio of the time-varying hybrid fiber-THz-fiber channel, is proposed. Based on the investigations of the aforementioned key techniques, we, for the first time, realize the photonics-assisted record-high 100/200 Gb E real-time THz wireless transmission at 360–430 GHz band, the capacity of which is 10–20 times higher than that of *** proposed fiber-THz-fiber architecture can realize the smooth conversion between high-speed THz and lightwave signals. Moreover, the architecture can significantly reduce the research difficulty and development cost, thereby considerably accelerating the commercialization of 6G THz technology by thoroughly reusing commercial digital coherent optical module(DCO) modules, which are compatible with the physical layer transmission protocols, such as IEEE 802.3 and ITU-T G.798. Finally, this study also introduces some potential directions of

关键词： 6G THz wireless communication optical fiber communication seamlessly converged architecture real-time communication

Simplified single-end Rayleigh and Brillouin hybrid distributed fiber-optic sensing system

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) 2023年第2期66卷 291-292页

作者： Linjing HUANG Zuyuan HE Xinyu FAN State key Laboratory of Advanced Optical Communication Systems and Networks Department of Electronic EngineeringShanghai Jiao Tong University Peng Cheng Laboratory

Dear editor,Distributed fiber-optic sensing(DFOS) technology has been widespread applications in many fields. It is mainly realized by analyzing three kinds of backscattering in optical fibers: Rayleigh, Brillouin, and Raman [1]. Conventionally,the sensing system can only measure one type of parameter:a static one like temperature and strain or a dynamic one like vibration. However, for several applications, such as railway safety monitoring systems, both static and dynamic parameters must be measured simultaneously not only for early warning applications but also for collecting static signals to perform long-term analysis.

关键词： Hybrid systems

A Golden Decade of Polar Codes:From Basic Principle to 5G Applications

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

China Communications 2023年第2期20卷 94-121页

作者： Kai Niu Ping Zhang Jincheng Dai Zhongwei Si Chao Dong The key Laboratory of Universal Wireless Communications Ministry of EducationBeijing University of Posts and TelecommunicationsBeijing 100876China The State key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China Department of Broadband Communication Peng Cheng LaboratoryShenzhen 518055China

After the pursuit of seventy years,the invention of polar codes indicates that we have found the first capacity-achieving coding with low complexity construction and decoding,which is the great breakthrough of the coding theory in the past two *** this survey,we retrospect the history of polar codes and summarize the advancement in the past ten ***,the primary principle of channel polarization is investigated such that the basic construction,coding method and the classic successive cancellation(SC)decoding are ***,in order to improve the performance of the finite code length,we introduce the guiding principle and conclude five design criteria for the construction,design and implementation of the polar code in the practical communication system based on the exemplar schemes in the ***,we explain the design principle behind the concatenated coding and rate matching of polar codes in 5G wireless ***,the improved SC decoding algorithms,such as SC list(SCL)decoding and SC stack(SCS)decoding etc.,are investigated and ***,the research prospects of polar codes for the future 6G communication system are explored,including the optimization of short polar codes,coding construction in fading channels,polar coded modulation and HARQ,and the polar coded transmission,namely polar ***,as a new coding methodology,polar codes will shine a light on communication theory and unveil a revolution in transmission technology.

关键词： polar codes channel polarization successive cancellation decoding polar coded modulation polar processing

Newton design:designing CNNs with the family of Newton's methods

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第6期66卷 122-137页

作者： Zhengyang SHEN Yibo YANG Qi SHE Changhu WANG Jinwen MA Zhouchen LIN School of Mathematical Sciences Peking University Bytedance AI Lab JD Explore Academy key Laboratory of Machine Perception School of Intelligence Science and TechnologyPeking University Pazhou Lab

Nowadays, convolutional neural networks(CNNs)have led the developments of machine ***, most CNN architectures are obtained by manual design, which is empirical, time-consuming, and non-transparent. In this paper, we aim at offering better insight into CNN models from the perspective of optimization theory. We propose a unified framework for understanding and designing CNN architectures with the family of Newton’s methods, which is referred to as Newton design. Specifically, we observe that the standard feedforward CNN model(PlainNet)solves an optimization problem via a kind of quasi-Newton method. Interestingly, residual network(ResNet)can also be derived if we use a more general quasi-Newton method to solve this problem. Based on the above observations, we solve this problem via a better method,the Newton-conjugate-gradient(Newton-CG)method, which inspires Newton-CGNet. In the network design,we translate binary-value terms in the optimization schemes to dropout layers, so dropout modules naturally appear in the derived CNN structures with specific locations, rather than being an empirical training *** experiments on image classification and text categorization tasks verify that Newton-CGNets perform very competitively. Particularly, Newton-CGNets surpass their counterparts ResNets by over 4% on CIFAR-10 and over 10% on CIFAR-100, respectively.

关键词： CNN dropout optimization method network design Newton's method

同方期刊数据库