在算法上

Firstly, some basic concepts about ECC are introduced;then the scalar multiplication under affine coordinate is discussed, we make a research and analysis of the side channel attack of scalar multiplication used in portable device, give out a new scalar multiplication algorithm which has the advantage of efficiency over other protected algorithm; and an improvement of the standard scalar multiplication algorithm proposed by IEEE1363 is made with the efficiency increasing by about 10%; at last, we deal with the using of complex multiplication in scalar multiplication algorithm, we generalize the methods and deduce a complete computing procedure, with proposing a new method which used several different fast endomorphism ,we give out an approach to boost the scalar multiplication with fast endomorphism.

本文首先介绍了椭圆曲线密码的有关基本概念；其次介绍了椭圆曲线上点的标量乘法在仿射坐标下的计算，对一般便携设备上的椭圆曲线点的标量乘法的边信道攻击做了研究与分析，给出了一种在效率上优于其它可抵抗边信道攻击的标量乘法的新算法，并对IEEE P1363 标准给出的标量乘法做了改进，使得标量乘法的运算效率提高了近10%；最后介绍了利用复乘计算标量乘法的方法，对已有的方法进行总结归纳，得出完整的计算过程，并针对两类超奇异椭圆曲线给出了一种利用多个可快速计算的复乘的标量乘法，得出一类普遍的结果，并给出了进一步用复乘加快标量乘法的思路。

In this dissertation we propose a set of fast algorithms for computing the DGFT (Discrete Generalized Fourier Transforms) over the hexagon domains and the parallel dodecahedron domains and give the parallel implementation on Linux clusters.

本文在理论上的主要贡献就是利用多色排序、区域分解和多重网格的思想，设计了二维三方向平行六边形区域和三维四方向平行十二面体区域上广义离散傅立叶变换的快速算法并给出它在PC Linux机群上的并行实现。

It not only supports Hash Operation on the first Chinese character of word,but ...

理论分析和实验结果表明，算法在速度和效率比现有的几种分词算法上有所提高。

Based on the idea of mapping the loops onto the reconfigurable array can improve the performance, the paper proposes a heuristic hardware-software partition algorithm which partitions the big loop into two parts: one is in the processor and another in the array. The objective of the algorithm is to minimize the communication of the two parts.

针对较大循环在可重构处理器上的映射问题提出了一种启发式的算法，将循环划分为在处理器上执行的软件部分和在可重构阵列上执行的硬件部分，并且使两者之间的数据传输量最小。

Conceptually, we present the principle of optimality for defuzzification and the optimal defuzzification mapping; methodologically, we convert the problem of finding the optimal defuzzification mapping into the one of finding the optimal parameters, suggesting a specific method for its realization—POD method; operationally, we advance that the learning of the parameters for defuzzification can be realized by the learning process using extended Kalman filter.

在概念上，提出了解模糊的优化原理和最优解模糊映射；在方法上，将求最优解模糊映射问题转化为一个参数优化问题，并提出了一个具体实现方法——POD方法；在实际操作上，我们提出可用基于增广Kalman滤波的参数学习算法来完成解模糊过程参数的学习。

In addition, the software about optical image system and some functional enhancement have been briefly introduced. The changes of the planar and spatial practical image point and ideal image point have been discussed when the vision field angle, object distance and range of refractive index have changed. Some summarizations have been put forward. Some results have been deduced and given qualitative explain.

并简单的介绍了由光学成像算法实现的软件以及对该算法的一些功能的增强从视场角、物距和折射率变化幅度三个方面讨论其分别发生变化时在平面上和空间中实际像点与理想像点的偏差的变化情况，并总结其变化规律，对其中的部分规律进行了简单的推倒和定性的说明。

The paper provided a new digital signature scheme according to the characteristics of factoring polynomials over a large infinite field and quadratic residue.

依据有限域中因式分解和二次剩余的特性，得到一种在GF上有效求解二次模p方程的算法，并根据该算法提出一种数字签名方案。

In the computer simulation aspect, following work has been down:At first, the Wolfgang Schreiner's model is employed and extended to a new computer arteriole optimal growing model. Many new algorithms are described, such as signal tracing, resistance calculating, resistance balancing and flow balancing. With these algorithms and methods, the arteriole networks can be grown successfully on microcomputer.Secondly, a growing-network ordering method, which is called auto-centripetal ordering method of the growing networks, can be employed to order the growing networks to calculate the average parameters of the networks in level series.

在计算机模型及仿真方面：首先本文改进、推广了WolfgangSchreiner模型，提出小动脉计算机优化生长模型，以及信号灯追踪、计算网络阻力、阻力平衡和流量平衡新算法，利用这种模型和算法成功地在微机上生长出了小动脉网络；第二，本文提出生长网络自动向心分级法，对生长网络进行了自动分级，并将网络参数的分级统计结果与A。

Just in this time, the deficiency of classic theory and algorithms of inductive learning were discovered that they had not put enough consideration to the predicted accuracy whereas emphasize the method of succinctness excessively.

正是在这一时期，人们发现了经典的归纳学习理论和算法的严重不足，即经典的归纳学习的理论和算法忽视的判决精度问题，而过分地强调了学习结果在概念上的简洁性。

This paper introduces the software structure and real time process principle of real time NC system, according to the milling process character of NC system and the geometrical modeling method of computer simulation, presents a technique of subdividing roughcasts and data structures which applies to the computer simulation of the NC milling process; and in order to quicken the operating rate and enhance the reality of computer simulation, puts forward a logic algorithm of the 3D milling process of real time NC system which can display the effects of the computer simulation of 3D milling process of real time NC system dynamicly and efficiently in PC, such as the movement of tools, the elimination of materials, the shape changing of roughcasts.

刘景能，蒙艳玫（广西大学，530004）摘要：本文介绍了实时数控系统软件结构与实时工作原理，根据数控系统铣削加工的特点和计算机仿真几何建模方法，给出了适用于数控铣削加工仿真的毛坯离散细分方法和数据结构；为了加快处理速度和增强仿真效果真实感，提出了一种实时数控系统三维加工仿真的逻辑设计算法，即在单CPU机器单线程程序中使用普通软件循环和定时器模拟实现多线程，该算法可以在微机上动态显示高质量的铣削加工过程刀具的运动，材料的去除和毛坯形状的变化等数控加工三维仿真效果。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。