In this paper, we study the boundary-value problem for mixed type equation with singular coefficient. We prove the unique solvability of the mentioned problem with the help of the extremum principle. The proof of the ...In this paper, we study the boundary-value problem for mixed type equation with singular coefficient. We prove the unique solvability of the mentioned problem with the help of the extremum principle. The proof of the existence is based on the theory of singular integral equations, Wiener-Hopf equations and Fredholm integral equations.展开更多
随着航天遥感领域高分辨率时间延迟积分成像中行频的提高,在一个积分时间内入瞳能量逐渐减少,在弱光条件下成像品质下降,需要采用增大积分级数的方法弥补能量的不足。现有传统数字域累加势必会导致叠加引入过多噪声,而模拟域又受到器件...随着航天遥感领域高分辨率时间延迟积分成像中行频的提高,在一个积分时间内入瞳能量逐渐减少,在弱光条件下成像品质下降,需要采用增大积分级数的方法弥补能量的不足。现有传统数字域累加势必会导致叠加引入过多噪声,而模拟域又受到器件工艺的影响无法实现大级数累加。文章首先提出了一种基于电荷域和数字域混合的累加方式,并阐述了具体成像方案。同时,结合相机在轨成像方式,对混合域累加方式下的信噪比和调制传递函数(Modulation Transfer Function,MTF)进行了理论推导。文章最后搭建实验环境对混合域累加方式下信噪比和MTF进行测试,并与传统电荷域进行比较,验证了混合域累加理论的正确性以及实现方法的可行性。根据分析与验证,文章所提方法有效的解决了单纯电荷域与数字域的主要瓶颈,为超高分辨率成像领域提供了有效的解决方案。展开更多
Entity matching is a fundamental problem of data integration.It groups records according to underlying real-world entities.There is a growing trend of entity matching via deep learning techniques.We design mixed hiera...Entity matching is a fundamental problem of data integration.It groups records according to underlying real-world entities.There is a growing trend of entity matching via deep learning techniques.We design mixed hierarchical deep neural networks(MHN)for entity matching,exploiting semantics from different abstract levels in the record internal hierarchy.A family of attention mechanisms is utilized in different periods of entity matching.Self-attention focuses on internal dependency,inter-attention targets at alignments,and multi-perspective weight attention is devoted to importance discrimination.Especially,hybrid soft token alignment is proposed to address corrupted data.Attribute order is for the first time considered in deep entity matching.Then,to reduce utilization of labeled training data,we propose an adversarial domain adaption approach(DA-MHN)to transfer matching knowledge between different entity matching tasks by maximizing classifier discrepancy.Finally,we conduct comprehensive experimental evaluations on 10 datasets(seven for MHN and three for DA-MHN),which illustrate our two proposed approaches1 superiorities.MHN apparently outperforms previous studies in accuracy,and also each component of MHN is tested.DA-MHN greatly surpasses existing studies in transferability.展开更多
Based on the dipole source method, all components of the Green＇s functions in spectral domain are restructured concisely by four basis functions, and in terms of the two-level discrete complex image method （DCIM） w...Based on the dipole source method, all components of the Green＇s functions in spectral domain are restructured concisely by four basis functions, and in terms of the two-level discrete complex image method （DCIM） with the high order Sommerfeld identities, an efficient algorithm for closed-form Green＇s functions in spatial domain in multilayered media is presented. This new work enjoys the advantages of the surface wave pole extraction directly carried out by the generalized integral path without troubles of that all components of Green＇s function in spectral domain should be reformed respectively in transmission line network analogy, and then the Green＇s functions for mixed-potential integral equation （MPIE） analysis in both near-field and far-field in multilayered media are obtained. In addition, the curl operator for coupled field in MPIE is avoided conveniently. It is especially applicable and useful to characterize the electromagnetic scattering by, and radiation in the presence of, the electrically large 3-D objects in multilayered media. The numerical results of the S-parameters of a microstrip periodic bandgap （PBG） filter, the radar cross section （RCS） of a large microstrip antenna array, the characteristics of scattering, and radiation from the three-dimensional （3-D） targets in multilayered media are obtained, to demonstrate better effectiveness and accuracy of this technique.展开更多
文摘In this paper, we study the boundary-value problem for mixed type equation with singular coefficient. We prove the unique solvability of the mentioned problem with the help of the extremum principle. The proof of the existence is based on the theory of singular integral equations, Wiener-Hopf equations and Fredholm integral equations.
文摘随着航天遥感领域高分辨率时间延迟积分成像中行频的提高,在一个积分时间内入瞳能量逐渐减少,在弱光条件下成像品质下降,需要采用增大积分级数的方法弥补能量的不足。现有传统数字域累加势必会导致叠加引入过多噪声,而模拟域又受到器件工艺的影响无法实现大级数累加。文章首先提出了一种基于电荷域和数字域混合的累加方式,并阐述了具体成像方案。同时,结合相机在轨成像方式,对混合域累加方式下的信噪比和调制传递函数(Modulation Transfer Function,MTF)进行了理论推导。文章最后搭建实验环境对混合域累加方式下信噪比和MTF进行测试,并与传统电荷域进行比较,验证了混合域累加理论的正确性以及实现方法的可行性。根据分析与验证,文章所提方法有效的解决了单纯电荷域与数字域的主要瓶颈,为超高分辨率成像领域提供了有效的解决方案。
基金the National Natural Science Foundation of China under Grant Nos.62002262,61672142,61602103,62072086 and 62072084the National Key Research and Development Project of China under Grant No.2018YFB1003404.
文摘Entity matching is a fundamental problem of data integration.It groups records according to underlying real-world entities.There is a growing trend of entity matching via deep learning techniques.We design mixed hierarchical deep neural networks(MHN)for entity matching,exploiting semantics from different abstract levels in the record internal hierarchy.A family of attention mechanisms is utilized in different periods of entity matching.Self-attention focuses on internal dependency,inter-attention targets at alignments,and multi-perspective weight attention is devoted to importance discrimination.Especially,hybrid soft token alignment is proposed to address corrupted data.Attribute order is for the first time considered in deep entity matching.Then,to reduce utilization of labeled training data,we propose an adversarial domain adaption approach(DA-MHN)to transfer matching knowledge between different entity matching tasks by maximizing classifier discrepancy.Finally,we conduct comprehensive experimental evaluations on 10 datasets(seven for MHN and three for DA-MHN),which illustrate our two proposed approaches1 superiorities.MHN apparently outperforms previous studies in accuracy,and also each component of MHN is tested.DA-MHN greatly surpasses existing studies in transferability.
基金the National Natural Science Foundation of China (Grant No. 60371020)National Defense Pre-research Foundation of China (Grant No. 9140a03020206dz0112)
文摘Based on the dipole source method, all components of the Green＇s functions in spectral domain are restructured concisely by four basis functions, and in terms of the two-level discrete complex image method （DCIM） with the high order Sommerfeld identities, an efficient algorithm for closed-form Green＇s functions in spatial domain in multilayered media is presented. This new work enjoys the advantages of the surface wave pole extraction directly carried out by the generalized integral path without troubles of that all components of Green＇s function in spectral domain should be reformed respectively in transmission line network analogy, and then the Green＇s functions for mixed-potential integral equation （MPIE） analysis in both near-field and far-field in multilayered media are obtained. In addition, the curl operator for coupled field in MPIE is avoided conveniently. It is especially applicable and useful to characterize the electromagnetic scattering by, and radiation in the presence of, the electrically large 3-D objects in multilayered media. The numerical results of the S-parameters of a microstrip periodic bandgap （PBG） filter, the radar cross section （RCS） of a large microstrip antenna array, the characteristics of scattering, and radiation from the three-dimensional （3-D） targets in multilayered media are obtained, to demonstrate better effectiveness and accuracy of this technique.
