Finding any given 2‐factor in sparse pseudorandom graphs efficiently

Given an $n$‐vertex pseudorandom graph $G$ and an $n$‐vertex graph $H$ with maximum degree at most two, we wish to find a copy of $H$ in $G$, that is, an embedding $\phi :V\left(H\right)\to V\left(G\right)$ so that $\phi \left(u\right)\phi \left(v\right)\in E\left(G\right)$ for all $uv\in E\left(H\right)$. Particular instances of this problem include finding a triangle‐factor and finding a Hamilton cycle in $G$. Here, we provide a deterministic polynomial time algorithm that finds a given $H$ in any suitably pseudorandom graph $G$. The pseudorandom graphs we consider are $\left(p,\lambda \right)$‐bijumbled graphs of minimum degree which is a constant proportion of the average degree, that is, $\mathrm{\Omega }\left(pn\right)$. A $\left(p,\lambda \right)$‐bijumbled graph is characterised through the discrepancy property: $|e\left(A,B\right)?p|A||B||<\lambda \sqrt{|A||B|}$ for any two sets of vertices $A$ and $B$. Our condition $\lambda =O(p^{2}n/\log n)$ on bijumbledness is within a log factor from being tight and provides a positive answer to a recent question of Nenadov. We combine novel variants of the absorption‐reservoir method, a powerful tool from extremal graph theory and random graphs. Our approach builds on our previous work, incorporating the work of Nenadov, together with additional ideas and simplifications.     

p-Adic函数空间上Hardy型算子的多线性交换子估计

在p-adic域上研究分数次Hardy型算子与CMO(Q_p~n)函数生成的多线性交换子,建立了交换子在Lebesgue空间和Herz空间上的有界性.对Hardy算子的多线性交换子也得到了相应的结果.     

1,3,5-三(羧基甲氧基)苯镍配位聚合物的合成、晶体结构及性质研究

采用溶剂热法,1,3,5-三(羧基甲氧基)苯为定向配体和乙酸镍反应构筑了一个新型的金属配位聚合物[Ni(TB)₂(H₂O)₂]_n·2H₂O,其中H₃TB=1,3,5-三(羧基甲氧基)苯,通过元素分析、IR及X射线单晶衍射对配合物结构进行表征,并研究其荧光性质、热稳定性及Hirshfeld表面作用力。单晶结构分析表明,该配合物属于三斜晶系,空间群$P \overline{1}$,配合物中心离子Ni(Ⅱ)分别与来自两个水分子上的氧原子及四个不同1,3,5-三(羧基甲氧基)苯配体的羧酸氧原子配位,形成六配位的NiO₆八面体构型,并通过与1,3,5-三(羧基甲氧基)苯配体的氧原子配位不断延伸形成具有孔洞结构的一维链状构型。配合物具有良好的荧光性能和热稳定性。Hirshfeld表面作用分析表明配合物分子中O…H/H…O作用占主导且占比为39.0%,而H…H的作用力占比为25.9%,O…O的作用力占比为13.6%。     

Terahertz generation in quantum cascade structures by two-color deriving fields: An approach to reach inversion population via optical pumping

In this paper, a quantum cascade laser (QCL) design is proposed based on GaAs/AlGaAs material system, which simultaneously operates at three widely separated wavelengths (${\lambda }_1=11.1\mu m,{\lambda }_2=14.1\mu m$ and ${\lambda }_{THz}=60\mu m$). In the design, all the wavelength radiations are achieved by the engineering of the electronic spectrum via the quantum-well widths and the applied electric field in a single active region within a same waveguide. The mid-infrared (mid-IR) wavelengths are obtained by adoption a dual-upper-state active region, and the proposed design aims to use both the mid-IR radiations as the coherent deriving fields to populate the upper THz lasing state to aid the THz-laser population inversion via optical pumping instead of direct electrical injection. A detailed analysis of electronic transport in the structure is carried out using a multi-level rate-equation model. The results show that the proposed structure offers an alternative approach to room temperature THz generation in QCLs.     

Acidity Modification of ZSM-5 for Methane Conversion in Co-feeding Method with MTA Reaction

Acidity plays a vital role in methane conversion by co-feeding method, which is one of the best strategies to improve the utilization and gentle the reaction conditions of methane. In this work, Zn, Ni, Mo, La, Ga, Fe and Co-impregnated ZSM-5 zeolites have been prepared with the same substitutions to variate the acidities and tested in co-aromatization of methanol with methane. It is demonstrated that the new medium-strong acid sites formed by metal and strong acid sites are the key role to activate methane in co-reaction. Zn-modified ZSM-5 catalyst is preferred to exhibit the best methane conversion of 12%, whose aromatic selectivity increases from 27.2% to 52.2% compared with that of HZSM-5. Besides, the addition of methane further improves the production of high-valued aromatics compared with methanol to aromatics (MTA) reaction.     

Coupling characteristics of laterally coupled gratings with slots

Laterally-coupled ridge-waveguide distributed feedback lasers fabricated without epitaxial regrowth steps have the advantages of process simplification and low cost. We present a laterally coupled grating with slots. The slots etched between the ridge and grating area are designed to suppress the lateral diffusion of carriers and to reduce the influence of the aspect-ratio-dependent-etching effect on the grating morphology in the etching process. Moreover, the grating height in this structure can be decreased to lower the aspect ratio significantly, which is advantageous over the conventional laterally coupled ridge waveguide gratings. The effects of five main structural parameters on the coupling characteristics of gratings are studied by MODE Solutions. It is found that varying the lateral width of the grating can be used as an effective way to tune the coupling strength; narrow slots (100 nm and 300 nm) and wide ridge (2 μm-4 μm) promote the stability of grating coupling coefficient and device performance. It is important to note that the grating bottom should be fabricated precisely. The comparative study of carrier distribution and mode field distribution shows that the introduction of narrow slots can strengthen the competitive advantage and stability of the fundamental mode.     

Sentinel-2卫星的多光谱轻量级船舶目标检测算法

近年来深度卷积神经网络在可见光船舶检测方面取得了显著的进展，然而，大多数相关研究是通过改进大型的网络结构来提高检测性能，因此加大了对更高计算机性能的需求。此外，可见光图像难以在云、雾、海杂波、黑夜等复杂场景检测到船舶。针对以上问题，提出了一种融合红（red, R）、绿（green, G）、蓝（blue, B）和近红外（NIR）4个波段光谱信息的由粗到精细的轻量型船舶检测算法。与现有的方法中根据光谱特性利用水体检测算法提取水体区域不同之处是该算法是利用改进的水体检测算法来提取船舶候选区域。为获取更准确的候选区域，对船舶、厚云、薄云、平静海面、杂波海面5种场景中4个波段的像素值进行了统计分析，选取近红外大于阈值作为辅助判断，并以其中心点获取候选区域32×32大小的切片，并对切片进行非极大值抑制，由此获得了船舶粗检测结果。随后构建了轻量级LSGFNet网络对船舶候选区域切片进行精细识别。构建的网络融合了1×1卷积提取的波谱特征与3×3的提取几何特征，为防止光谱特征与几何特征的信息在融合时“信息不流通”，在LSGFNet网络中引入了ShuffleNet中的通道打乱机制，并减小了模型结构，与典型的轻量级网络相比具有更好的效果且模型较小。最后，利用Sentinel-2卫星多光谱10 m分辨率数据构建了512×512大小的1 120组数据进行粗检测，以及32×32大小的6 014组数据进行精细网络训练，其中候选区域粗提取的查全率为98.99%，精细识别网络精确度为96.04%，不同场景下的平均精确度为92.98%。实验表明该算法在抑制云层、海浪杂波等干扰的复杂背景下具有较高的检测效率，且训练时间短、计算机性能需求低。