ID 原文 译文
2493 在搜狐新闻标题和微博短文本数据集上进行相关实验,结果表明 STOCIRNMF 不仅比现有代表性算法具有更好的聚类性能,而且能够有效对微博话题进行在线检测。 We conduct extensiveexperiments on real Sohu news titles and Weibo datasets. The results show that STOCIRNMF not only has better perform-ance of short text clustering than some representative algorithms, but also is very effective to detect micro blog's topics on-line.
2494 提出了一种新型平面三频带通滤波器,该滤波器由一个加载短路枝节的阶梯阻抗谐振器,一对加载开路枝节的背靠背 E 型谐振器,以及包含源负载直接耦合的馈电结构组成。 A new microstrip tri-band bandpass filter is presented. The filter is composed of a stepped-impedance re-sonator loaded with short stub, a pair of E-shape resonators loaded with open stub, and the external coupling structure withinput /output direct-coupling.
2495 所采用的枝节加载谐振器的多模工作特性使滤波器的体积大大减小,同时每个通带的位置及其耦合特性都能够独立调谐。 Due to the multi-mode characteristics of the resonators, the filter size can be significantly mini-aturized, and each passband can be separately controlled.
2496 另外,通过源负载直接耦合引入通带两侧的传输零点,实现了滤波器良好的频率选择性。 In addition, the proposed filter can generate six transmission zeroson both sides of each passband, resulting in improved frequency selectivity.
2497 最后设计并加工了一款高选择性小型化三频带通滤波器,其三个通带的中心频率分别为 2.0GHz,3.95GHz 6.35GHz, Such a demonstrative tri-band bandpass filterwith center frequencies of 2. 0GHz, 3. 95GHz and 6. 35GHz were designed and fabricated.
2498 插入损耗均小于 2.5dB,带内回波损耗均优于 14dB,实验结果与仿真结果吻合良好。 The measured results show that the insertion loss is less than 2. 5dB, and the return loss is better than 14 dB for each passband, which is in agreement with the simulated ones.
2499 穿墙雷达成像中,墙体反射杂波干扰严重,严重影响目标成像效果。 In Through-the-Wall Imaging (TWI), the target image is often buried in the wall reflections and otherclutter waves, and the target image cannot be obtained.
2500 子空间技术对回波信号矩阵进行奇异值分解后去除墙体子空间,可以有效的抑制墙体杂波干扰,在穿墙雷达成像中具有广泛的应用。 Subspace technique uses the received echo matrix Singular Value De-composition (SVD)to remove the wall subspace, it is a mature technique and has broad applications in TWI.
2501 该文针对子空间技术中墙体与目标子空间的划分这一难题,提出一种基于奇异向量平稳度分类的墙体子空间提取技术。 In order to distinguish the wall subspace and the target subspace, this paper presents a new wall subspace extractive method based on the stationarity of left-singular vector.
2502 该方法利用墙体回波信号的相关特征,根据奇异值分解后各个左奇异向量的平稳程度来精确划分墙体与目标子空间。 SVD is applied to the echo matrix and the left-singular vectors are used to determine the subspace belonging to the wall or the target.