类域

The general principle of RR STAP is addressed in this paper with the result that the output signal to clutter plus noise ratio will be probably near as or even better than that of full rank STAP. Thereby, the final detection performance of practical airborne phased array radar won't necessarily be degraded when rank reduction is operated properly. There exists two basic types of adaptive processor, one is Direct Form Processor and the other is Generalized Sidelobe Canceler .

降维可以在空域、时域或时空联合域进行，无论在哪个域内进行，都可以等效成对全空时数据矢量施以一个降维变换矩阵Q〓，如果Q〓与数据的特征结构无关，那么这类降维处理器的结构预先也是固定的，如果Q〓与数据的特征结构有关，即STAP在特征子空间内进行，那么所选择的子空间不同，STAP性能也不一样。

The original local scope (the one in effect just before the class definition was entered) is reinstated, and the class object is bound here to the class name given in the class definition header.

原始的局部作用域（类定义引入之前生效的那个）得到恢复，类对象在这里绑定到类定义头部的类名（例子中是 ClassName ）。

The structure theorem is obtained by defining "ternary ordered partition method" which is used to choose the test-sets of the first $k$ tests, and by defining two operations and two relations on feedback sequences; The isomorphism classification theorem is strictly proved by virtue of the structure theorem, thus the complicated problem of discussing $3^k$ possible search domains is reduced to the relatively easy problem on $k+1$ representative search domains; By the techniques of "super-coin construction method" and "image-union method", the search problem on $k+1$ representative search domains is reduced to the simple verification of several conditions on finite space of small cardinality.

通过引入用于选取前$k$次试验的试验集的"三元有序剖分法"以及关于反馈序列的两种运算和两种关系，得到了结构定理；基于结构定理严格地证明了同构类定理，从而将必须针对$3^k$个搜索域的复杂问题转变为讨论$k+1$个具代表性的搜索域的相对简单的问题；通过"构造超币法"和"并像法"等技术手段，将关于$k+1$个具代表性的搜索域的搜索问题转化为简单的关于势非常小的有限空间的几个条件的验证问题。

In order to overcome the handicap to extract line-segments from natural hand-printed Chinese characters, the extension is made to other structural features, and the generalized connected background region is defined, and its extraction and recognition algorithm are developed. This recognition algorithm is based on matrix singular decomposition and invariant transformation theory.

针对从自然手写体汉字中难以提取到笔段特征的难题，对汉字的其它结构类特征进行了扩充，提出了汉字的广义背景连通域的概念，并给出了提取广义背景连通域特征的算法以及基于矩阵奇异值分解与不变性变换的广义背景连通域特征的匹配算法。

School of Information Engineering,Zhengzhou University,Zhengzhou 450052,ChinaA novel chirp-typed watermarking scheme which uses the properties of index additivity and orthonormality of fractional Fourier transform to embed watermark in the different fractional Fourier domain is presented.

提出了一种新的分数阶Fourier域Chirp类数字水印方案，该方案利用分数阶Fourier变换基函数的正交性和旋转相加性，在不同的分数阶Fourier域嵌入Chirp水印，并利用分数阶Fourier域Chirp信号的聚集性进行盲检测。

The method includes the following steps: firstly, a spherical neighborhood of present sample points is established by using the geometric spherical-modelling theory and all the sample points contained in the spherical neighborhood are adopted as candidate neighbor points, thus not only preserving the effectivity of the dimension reduction capability when data sets are sparse but also getting the advantages of low-sensitivity to isolated points and good stability of the preserved topological structure; then a data relevance matrix more matching semantics can be obtained by relevance measurement based on route clusters to update the candidate neighbor points in the spherical neighborhood and optimize the regular neighborhood space of the present sample points, thus improving the phenomenon that the dimension reduction of sample sets provided with folded curved faces is apt to suffer the integrated-structure distortion in case of heterogeneous data distribution.

首先利用几何开球原理建立当前样本点的球状邻域，将包含在球状邻域内的所有样本点作为候选近邻点，不但能够保持在数据集稀疏情况下的降维性能的有效性，而且具有对孤立点敏感性不高、保留拓扑结构稳定性好的优点。然后利用基于路径聚类的相关性度量得到更符合语义的数据相关性矩阵，用来对球状邻域内的候选近邻点进行更新，优化当前样本点的规则邻域空间，改善了当数据不均匀分布时在带有折叠弯曲面的样本集上降维容易出现整体结构扭曲的现象。

In this paper,the author give the spectra characterizations of nonsigular feedback polynomials and local permutation polynomial s over finite fields and residue class rings,and simplify existing results over prime fields,and provide a new proof method for the spectra characterization of correlation immune functions over finite fields and residue class rings.

本文给出了有限域和剩余类环上非线性反馈移存器的非奇异反馈多项式及局部置换多项式的谱刻划，简化了素域上的现有结果，并对有限域上和剩余类环上相关免疫函数的谱特征给出了一个新的证明方法。

According to the analysis of features of the key outcrop sequence stratigraphy, build the sequence frame, through the section comparison, and the study of feature of the 5 key outcrops, then it can be concluded that, conglomerate on the denudation plane or the other contrastive sediment are the main features surrounding the bottom border of the sequence. In the foreland basin sediment environment, the transilient group of mud rock and sandstone are the main sediment features of low-stand system tract. Thick mud rock and sandstone interbedded is the chief component of transgressive system tract, sometimes together with thick turbidity current sandstone; High-stand system tract is composed by the combination of mud rock and sandstone interbedded, with the progradation as its main style. Maximum flooding surface Sediment are usually mudstone, and the thickness is different from each other.

依据关键露头剖面高精度层序特征分析，通过剖面对比建立高精度等时层序地层格架，通过对5条关键露头剖面的详细高精度层序特征分析，认识到：剥蚀面上的粗碎屑岩或与之可区域对比的其它沉积是层序底部边界附近的主要沉积特征；在盆地沉积环境中，砂泥岩发育，厚层、块状砂岩与泥质岩类的突变组合是低位体系域的主要沉积特征；湖进体系域以厚层泥岩夹砂岩为主，有时夹厚度较大的重力流砂体；高位体系域为砂泥互层沉积组合，以进积式叠加样式为主；最大湖泛面多为泥岩，但其厚薄不一。

Based on this, aim at the disadvantage on Wigner Distribution and Fractional Fourier Transform that can't restrain noise interference effectively when be used to detect Linear Frequency Modulated signals interfered by heavy noises, we proposed a improved WD. Simulation results show that this method can restrain noise interference effectively and improve the ability of LFM signals detection. On the other hand, in multi-component cases, WD suffers from the cross-terms, which may impair some of the auto-terms. In order to separate original multi-component signals into components and reconstruct them separately, we can use the method of optimal filtering in fractional Fourier domain. But this method also has a problem that if signals have aliasing in the fractional Fourier optimal domain as well, the reconstructed signals will be distorted. In this thesis, a novel method of quasi Fourier transform is proposed in order to solve the problem of multi-component signal reconstruction, and have good results.

在此基础上，针对Wigner分布和分数维傅里叶变换在处理高噪声背景下线性调频信号检测的问题时存在不能有效抑制噪声干扰的缺点，提出了改进Wigner分布的办法，该方法能够抑制噪声，提高信号的检测能力；针对Wigner分布在处理多分量非线性调频信号时存在干扰项的缺点，以及分数域最优滤波方法在处理多分量非线性调频信号分离重构问题时存在的局限性，该局限性表现为在对最优分数域上也有信号分量重叠的情况进行处理时重构信号发生畸变，提出了新颖的类傅里叶变换的方法，该方法对在最优分数域上也有信号分量重叠的情况也能处理，得到的重构信号失真很小。

Audio denoising is the pre-step in the audio processing,and wavelet domain denoising has been a main research area.In order to give people a summary knowledge of wavelet domain audio denoising,based on many literatures,this paper attempts to make a survey of wavelet domain audio denoising.First,it introduces wavelet transformation and the characteristics,and gives the optimal selecting and dealing criterion of wavelet domain denoising,the basis wavelet selection.Second,describes five classical methods of audio denoising,for example,threshold,multi-wavelet,wavelet packet,modulus maximum and scale correlation,then introduces their development.Third,introduces a new algorithm based on modulus maximum and scale trading theory.At the end,the future trend of wavelet domain audio denoising is pointed out.

音频降噪是音频处理前期的必要步骤，小波域降噪是其中一个重要的研究方向，为了使人们对小波域音频降噪有概括了解，在对小波域音频降噪相关文献进行分析和理解的前提下，首先介绍了小波变换及其特性，同时阐述了小波降噪的最优选取和处理准则，及对音频进行小波变换时小波基的选取原则；然后评述了用于音频降噪的阈值，多小波、小波包、模极大值和尺度相关法等5类方法及各自的发展；接着介绍在前人基础上，提出新的基于模极大值和尺度理论的降噪算法；最后探讨了小波域音频降噪的发展方向。

Finally, according to market conditions and market products this article paper analyzes the trends in the development of camera technology, and designs a color night vision camera.

最后根据市场情况和市面上产品的情况分析了摄像机技术的发展趋势，并设计了一款彩色夜视摄像机。

Only person height weeds and the fierce looks stone idles were there.

只有半人深的荒草和龇牙咧嘴的神像。