函数

Though comparing Canny operator and center B spline dyadic wavelet, the following conclusion is proven in this dissertation: a Center B spline function has tight support and Canny operator hasn't. b Center B spline function asymptotic convergence to Gaussian function and the derivative of Center B spline function asymptotic convergence to Canny operator. c The derivative of fourth order center spline B function is more suitable as a optimal edge detector than Canny operator. d Center B spline function can balance the smoothing and approximation of original data, and the fourth center B spline function is the only optimal solution of two order smoothing problem. e The error between the valve of time-frequency uncertainty of the fourth center B spline function and the lower bound of time-frequency uncertainty does not exceed 0.143% of the lower bound. f The derivative of center spline B function can construct a stability dyadic wavelet and can give a fast algorithm for multiscale edge detection, but Canny operator can do neither.

作者给出了Canny算子与中心B样条二进小波严格的比较证明，得出如下结论：a中心B样条函数具有紧支集，Canny算子不具有紧支集。b中心B样条函数的极限收敛于高斯函数，中心B样条函数的导数收敛于Canny算子。c四阶中心B样条函数的导数比Canny算子更接近最佳边缘检测滤波器。d中心B样条函数比高斯函数更能兼顾对原函数平滑和逼近的折中要求，并且四阶中心B样条函数是二阶逼近问题的唯一最优解。e四阶中心B样条函数的时频测不准关系值与时频测不准关系下界的逼近误差不超过0.143％。f中心B样条函数的导数可以构成稳定的二进小波，存在快速的多尺度算法；而Canny算子不构成稳定的二进小波，无法给出快速的多尺度算法。

Bring forward a kind of learning algorithm fitting on the base of spline function, because the spline function has good flexibility and quadratic smoothing property, the search problem of network weight functions training can be converted into the resoluting extreme value problem of the plurality function through using spline function to represent network input and link weight functions, thus the PNN can be trained with the help of the existing neural network learning algorithms;④build a kind of learning algorithm base on the alternate of Walsh function, the complexity of network computing can be decreased through using the complete orthogonality of Walsh function corollary.

应用结果表明该算法收敛速度快，稳定性好；③提出了一种基于样条函数拟和的学习算法，由于样条函数具有很好的柔韧性和二次光滑性，将网络输入函数和连接权函数用样条函数的形式表示，可把网络权函数训练的函数寻优问题转换为多元函数求极值的问题，从而可借助于现有的神经网络学习算法训练过程神经元网络；④建立了一种基于Walsh函数变换的学习算法，利用Walsh函数系的完备正交性，可大大降低网络计算的复杂度。

The main contributions of the second part of this dissertation are focused on the cryptographic properties of logical functions over finite field, with the help of the properties of trace functions, and that of p-polynomials, as well as the permutation theory over finite field: The new definition of Chrestenson linear spectrum is given and the relation between the new Chrestenson linear spectrum and the Chrestenson cyclic spectrum is presented, followed by the inverse formula of logical function over finite field; The distribution for linear structures of the logical functions over finite field is discussed and the complete construction of logical functions taking on all vectors as linear structures is suggested, which leads to the conception of the extended affine functions over finite field, whose cryptographic properties is similar to that of the affine functions over field GF (2) and prime field F〓; The relationship between the degeneration of logical functions and the linear structures, the degeneration of logical functions and the support of Chrestenson spectrum, as well as the relation between the nonlinearity and the linear structures are discussed; Using the relation of the logical functions over finite field and the vector logical functions over its prime field, we reveal the relationship between the perfect nonlinear functions over finite field and the vector generalized Bent functions over its prime field; The existence or not of the perfect nonlinear functions with any variables over any finite fields is offered, and some methods are proposed to construct the perfect nonlinear functions by using the balanced p-polynomials over finite field.

重新定义了有限域上逻辑函数的Chrestenson线性谱，考察了新定义的Chrestenson线性谱和原来的Chrestenson循环谱的关系，并利用一组对偶基给出了有限域上逻辑函数的反演公式；给出了有限域上随机变量联合分布的分解式，并利用随机变量联合分布的分解式对有限域上逻辑函数的密码性质进行了研究；给出了有限域上逻辑函数与相应素域上向量逻辑函数的关系，探讨了它们之间密码性质的联系，如平衡性，相关免疫性，扩散性，线性结构以及非线性度等；讨论了有限域上逻辑函数各类线性结构之间的关系，并给出了任意点都是线性结构的逻辑函数的全部构造，由此引出了有限域上的"泛仿射函数"的概念；考察了有限域上逻辑函数的退化性与线性结构的关系、退化性与Chrestenson谱支集的关系；给出了有限域逻辑函数非线性度的定义，利用有限域上逻辑函数的非线性度与相应素域上向量逻辑函数非线性度的关系，考察了有限域上逻辑函数的非线性度与线性结构的关系；利用有限域上逻辑函数与相应素域上向量逻辑函数的关系，揭示了有限域上的广义Bent函数与相应素域上的广义Bent函数的关系，以及有限域上的完全非线性函数与相应素域上向量广义Bent函数之间的关系；给出了任意有限域上任意n元完全非线性函数存在性与否的完整证明，并利用有限域上平衡的p-多项式的性质给出了有限域上完全非线性函数的一些基本构造方法。

This paper discusses such problems which are easily solved incorrectly in limitedteaching as, segment function is uncertain not to be elementary function, elementary function is un-certain to be continuous in its field of definition, unlimited variable is uncertain to be infinity, theinfinitely many infinitesimal is uncertain to be infinitesimal, the multiplication of two non- infinitesi-mal is uncertain not to be infinitesimal.

1分段函数不一定不是初等函数我们知道初等函数的定义是：常数和基本初等函数经过有限次的四则运算或复合，用一个解析式表示的函数叫初等函数。教材中的定义强调的"有限次"和"一个解析式"，使很多同学产生"非有限次"和"非一个解析式"的函数就不是初等函数的误解。例1f=ceox，s xx！0为分段函数，但也可以表示为f=e12(x-x2")+cos12(x+x2#"$)-1，故为初等函数。

Using the concept of Boolean functions and combinatorics theory comprehensively, we investigate the construction on annihilators of Boolean functions and the algebraic immunity of symmetric Boolean functions in cryptography:Firstly, we introduce two methods of constructing the annihilators of Boolean functions, Construction I makes annihilators based on the minor term expression of Boolean function, meanwhile we get a way to judge whether a Boolean function has low degree annihilators by feature matrix. In Construction II, we use the subfunctions to construct annihilators, we also apply Construction II to LILI-128 and Toyocrypt, and the attacking complexity is reduced greatly. We study the algebraic immunitiy of (5,1,3,12) rotation symmetric staturated best functions and a type of constructed functions, then we prove that a new class of functions are invariants of algebraic attacks, and this property is generalized in the end.Secondly, we present a construction on symmetric annihilators of symmetric Boolean functions.

本文主要利用布尔函数的相关概念并结合组合论的相关知识，对密码学中布尔函数的零化子构造问题以及对称布尔函数代数免疫性进行了研究，主要包括以下两方面的内容：首先，给出两种布尔函数零化子的构造方法，构造Ⅰ利用布尔函数的小项表示构造零化子，得到求布尔函数f代数次数≤d的零化子的算法，同时得到通过布尔函数的特征矩阵判断零化子的存在性：构造Ⅱ利用布尔函数退化后的子函数构造零化子，将此构造方法应用于LILI-128，Toyocrypt等流密码体制中，使得攻击的复杂度大大降低；通过研究(5，1，3，12)旋转对称饱和最优函数的代数免疫和一类构造函数的代数免疫，证明了一类函数为代数攻击不变量，并对此性质作了进一步推广。

The problem of multi-objective usually meet many conflict each other and can not use the objective of the same standard unit, then to employ membership function of fuzzy theory,at first each objective function to change fuzzy set and express with membership function, membership function include the maximum constraint, minimum constraint, the decrease function of monotonic property, to become optimal of the multi-objective function, but still to scanty of weighting value with regard to adjustment of objective function, thus this thesis proposes the method of combine orthogonal arrays and particle swarm optimization to solve the problem of multi-objective optimal power flow, each objective function separately add weighting value, to setting each weighting value of objective function in order to the result of anticipating.

多目标最佳化的问题通常会面临到许多互相冲突且不能用同ㄧ标准单位的目标，於是利用模糊理论中的归属函数，首先将各目标函数转换成模糊集合并以归属函数表示，归属函数包含了最大限制值、最小限制值、单调的递减函数所组成，将各目标函数利用归属函数表示，形成单一目标函数的最佳化，但是仍缺乏权重值对於目标函数的调整，於是本篇论文提出ㄧ种权重值设定与粒子群优演算法的方法去解决多目标最佳电力潮流的问题，将各目标函上分别加上权重值设定各目标函数的权重值，得到预期的效果。

The Production Function 生产函数 The production function shows the relationship between quantity of inputs used to make a good and the quantity of output of that good.

生产函数——用于生产一种物品的生产函数用于生产一种物品的投入量与该物品产量之间的关系。投入量与该物品产量之间的关系。

It introduces partial fractions of meromorphic functions, product developments of entire functions, Hadamard's theorem, Riemann Zeta functions, Poisson-Jensen's formula; elliptic functions, including simply periodic functions and doubly periodic functions; algebraic functions and algebroid functions, Riemann surface, Nevanlinna theory, including characteristic functions, the first and second fundamental theorems, growth orders, etc; complex differential equations and complex functional equations, etc.

具体为：亚纯函数的部分分式、整函数的无穷乘积展开、Hadamard定理、Riemann Zeta函数、Poisson-Jensen公式；椭圆函数，包括单周期函数、双周期函数；代数函数和代数体函数、Riemann曲面简介；Nevanlinna理论简介，包括特征函数、第一和第二基本定理、增长级等；复微分方程和复函数方程，等等。在教学内容上充分体现基础性、新颖性。

From idautils import * funclist=[0x004011F2, 0x00401200, 0x00401206] /*add all the other addresses here*//*在这里加上所有函数的起始地址*/ ea=ScreenEA for i in funclist: ea=i add_bpt(ea, 1, 4) bp=bpt_t get_bpt bp.flags ^=BPT_BRK bp.flags|=BPT_TRACE 下面给出我的一个实现: from idaapi import * from idautils import * funclist = [0x004011F2, 0x00401200, 0x00401206] ea=ScreenEA for line in funclist: ea = line add_bpt(ea,1,4) SetBptAttr For those cases where it's necessary to mark absolutely every function as a tracepoint, the IDA plug-in can be automated to find every function reference instead of doing it manually.

有了上面这个脚本之后，我们实际上就已经可以在每个函数的起始位置上都下一个 tracepoint 了。不过手工把每个函数的起始地址都写入上面那张表里还是比较枯燥的一个活，好在 IDA 插件里本身就提供一个能遍历所有函数的功能。我们完全可以编写一个能自动在每个函数的起始位置上都下上一个 tracepoint 的脚本出来，比如下面给出的这个脚本。

The purpose of this dissertation is to study a principle or method to describe and analyze the complex problems with high gradients arising from practical engineering. The main contents cover:1. The Bezier function was introduced to approximate high gradient functions, and a suitable coupled function was derived. Some detail numerical results show that the Bezier-based function is dominant over other combination function such as polynominal and trigonometric series, when to describe a high gradient function.

针对工程中大量存在的高梯度问题进行函数逼近的研究，引入Bezier函数对高梯度问题进行函数逼近，构成以Bezier函数为主的复合逼近函数，数值考证表明：与常规多项式插值函数的逼近及"幂／三角"复合函数的逼近相比，基于Bezier函数的逼近函数对高梯度问题的描述具有明显优势。

I didn't watch TV last night, because it .

昨晚我没有看电视，因为电视机坏了。

Since this year, in a lot of villages of Beijing, TV of elevator liquid crystal was removed.

今年以来，在北京的很多小区里，电梯液晶电视被撤了下来。