模重复平方算法的rho改进算法

利用循环二进制方法给出了适合大指数模乘运算的模重复平方算法的rho改进算法,以提高模幂乘法的计算速度。新算法的实质是一种指数约减算法,可以有效减少模重复平方算法中的模乘运算。通过实例计算表明,新算法可以极大地提高运算速度。     

二维主成分分析方法的推广及其在人脸识别中的应用

提出了分块二维主成分分析(分块2DPCA)的人脸识别方法。分块2DPCA方法先对图像矩阵进行分块,对分块得到的子图像矩阵直接进行鉴别分析。其特点是:能方便地降低鉴别特征的维数;可以完全避免使用矩阵的奇异值分解,特征抽取方便;与2DPCA方法相比,使用低维的鉴别特征矩阵,而达到较高(至少是不低)的正确识别率。此外,2DPCA是分块2DPCA的特例。在ORL和NUST603人脸库上的试验结果表明,所提出的方法在识别性能上优于2DPCA方法。     

基于进化硬件的自修复TMR系统设计及其可靠性分析

将进化硬件与传统TMR容错设计思想相结合,提出了一种具有在线自修复功能的自修复TMR系统设计方法。该系统具有多重容错和修复机制:总体采用TMR,可自动检测到故障模块;系统中每个模块均采用组件备份法,可通过组件切换法快速修复模块故障;而模块中每个组件也可通过进化进行修复。因而具有更强的容错能力和更高的可靠性。以具有片内三模冗余的2 bit乘法器为例进行了验证。最后,给出了该系统的可靠性模型,推出了可靠性计算公式,从理论上对该系统的可靠性进行了分析。结果表明:该系统能有效修复stuck-at故障,具有更长的使用寿命和更高的可靠性。     

Automated modeling of modular robotic configurations

This research presents an automated method to build kinematic and dynamic models for assembling modular components of modular robotic systems. By comparison with other approaches, the proposed method is applicable to any robotic configuration with serial, parallel, or hybrid structures. In addition, it is object oriented so that each modular component is an element with a submodel and the overall model can be assembled from submodels subject to the connection constraints.     

Autonomous reconfiguration of robot shape by using Q-learning

A modular robot can be built with a shape and function that matches the working environment. We developed a four-arm modular robot system which can be configured in a planar structure. A learning mechanism is incorporated in each module constituting the robot. We aim to control the overall shape of the robot by an accumulation of the autonomous actions resulting from the individual learning functions. Considering that the overall shape of a modular robot depends on the learning conditions in each module, this control method can be treated as a dispersion control learning method. The learning object is cooperative motion between adjacent modules. The learning process proceeds based on Q-learning by trial and error. We confirmed the effectiveness of the proposed technique by computer simulation.     

基于灵敏度分析的模块化回声状态网络修剪算法

针对回声状态网络（Echo state network,ESN）的结构设计问题,提出基于灵敏度分析的模块化回声状态网络修剪算法（Pruning algorithm for modular echo state network,PMESN）.该网络由相互独立的子储备池模块构成.首先利用矩阵的奇异值分解（Singular value decomposition,SVD）构造子储备池模块的权值矩阵,并利用分块对角阵原理生成储备池.然后利用子储备池模块输出和相应的输出层权值向量,定义学习残差对于子储备池模块的灵敏度以及网络规模适应度.利用灵敏度大小判断子储备池模块的贡献度,并根据网络规模适应度确定子储备池模块的个数,删除灵敏度低的子模块.在网络的修剪过程中,不需要缩放权值就可以保证网络的回声状态特性.实验结果说明,所提出的算法有效解决了ESN的网络结构设计问题,基本能够确定与样本数据相匹配的网络规模,具有较好的泛化能力和鲁棒性.     

Morphological approach for autonomous and adaptive systems based on self-reconfigurable modular agents

This paper describes a novel approach for managing self-organizing, distributed modular components in dynamically changing environments. The main concept is to fabricate a system which is composed of dynamically associated modular agents, that can migrate and reorganize by itself while the system is being executed. Association between modular agents can be varied and transmuted according to components’ own migration schemes including deployment based on biological processes. This paper presents a self-organizable architecture, which can reorganize and reconfigure a system based on modular agents. It is contrived through observation of biological phenomena, and implements a platform to host the architecture in dynamically changing environments. We draw several key features of the modular agents, describe the principles of the modular agent based self-organizable framework, and depict how the proposed framework satisfies the functional requirements of network applications, which are made of several agents. We also demonstrate the efficiency and scalability of the framework through examining some simulation results.     

动态图编辑器框架设计与实现

从具体的应用中抽象出适用于动态处理的图编辑器框架，并在此基础上利用JAVA技术构造动态图编辑器API。它的开放结构允许应用程序在运行时,使其一方面能够独立于应用程序，维护其自身的稳定性，另一方面又能够和应用程序有机地结合在一起。它的模块化结构为应用程序提供了不同层次的切入点。动态图编辑器框架作为可重用的设计，能够对相关领域图应用软件的研制起指导作用，在此基础上建立的图应用软件不但具有易扩展与维护的特点，而且可以在较短的时间内完成。     

改进的模块2DPCA人脸识别算法

提出一种改进的模块2DPCA人脸识别算法,即基于子距离的模块2DPCA人脸识别算法。该算法对图像进行分块,对每一子块独立地利用2DPCA进行处理,求出测试样本子块与训练样本对应子块间的子距离,将所有子距离相加得到测试样本与训练样本的距离,用最近距离分类器分类。在ORL人脸库上的实验结果表明,该算法在识别性能上优于普通的模块2DPCA算法和修正的模块2DPCA算法。     

聚乙烯装置模块多变量在线操作指导

提出了模块多变量在线操作指导系统的一般性框架结构，并设计了一个模块多变量在线操作指导系统．它能及时给出合适的线性低密度聚乙烯反应器催化剂操作量的指导值，预防反应器温度超高限运行；在稳定反应器的温度、生产负荷和床层重量方面取得了良好效果．     

基于250位模乘平台的Tate对最终模幂算法的改进

在只支持250 bits模乘的硬件平台上,实现457 bits的二元扩域Tate对Miller算法的（双线性对的一种）最终模幂运算。在计算过程中采用一种改进的Montgomery模乘算法和中国剩余定理算法。通过具体数据实现双线性对最终模幂的运算,使用数学软件Sage来验证这种改进方案的正确性。通过理论分析和数据计算可以证明使用该方案可实现457 bits最终模幂。     

An efficient common-multiplicand-multiplication method to the Montgomery algorithm for speeding up exponentiation

The modular exponentiation is a common operation for scrambling secret data and is used by several public-key cryptosystems, such as the RSA scheme and DSS digital signature scheme. However, the calculations involved in modular exponentiation are time-consuming especially when performed in software. In this paper, we propose an efficient CMM-MSD Montgomery algorithm by utilizing the Montgomery modular reduction method, common-multiplicand-multiplication (CMM) method, and minimal-signed-digit (MSD) recoding technique for fast modular exponentiation. By using the technique of recording the common signed-digit representations in the grouped substrings of exponent, our algorithm can improve the efficiency of both the original CMM exponentiation algorithm and the Montgomery multiplication algorithm. The fast modular exponentiation algorithm developed in this paper can be easily implemented in general signed-digit computing machine, and is therefore well suited for parallel implementation to fast evaluating modular exponentiation. Moreover, by using the proposed CMM-MSD Montgomery algorithm, on average the total number of single-precision multiplications can be reduced by about 38.9% and 26.68% as compared with Dusse-Kaliski’s Montgomery algorithm and Ha-Moon’s Montgomery algorithm, respectively.     

一种改进的模块PCA方法及其在人脸识别中的应用

提出了一种改进的模块PCA方法,即基于类内平均脸的分块PCA算法。该算法对每一类训练样本中每个训练样本的每一子块求类内平均脸,并用类内平均脸对训练样本类内的相应子块进行规范化处理,然后由所有规范化后的子块构成总体散布矩阵,从而得到最优投影矩阵;由训练集的全体子块的平均值对训练样本的子块和测试样本的子块进行规范化后投影到最优投影矩阵,得到识别特征;最后用最近距离分类器分类。在ORL人脸库上的试验结果表明,提出的方法在识别性能上明显优于普通模块PCA方法。     

A knowledge encapsulation approach to ontology modularization

The development of monolithic ontologies for complex domains may face various challenges in reasoning and implementation. The notion of modularity can be employed for developing more efficient ontologies, especially in distributed environments. In this paper, we introduce a framework for developing ontologies in a modular manner. We describe the interface-based modular ontology formalism, (IBF), which theoretically supports the framework. The main feature of the framework is its support for knowledge encapsulation, i.e., it allows ontologies to define their main content using well-defined interfaces, such that their knowledge bases can only be accessed by other ontologies through these interfaces. An important implication of the proposed framework is that ontology modules can be developed completely independent of each other’s signature and languages. Such modules are free to only utilize the required knowledge segments of the others. We also investigate the issues of inconsistency in the proposed modular ontology framework. We provide solutions for isolating inconsistent ontology modules from the other parts of a modular ontology and also resolve inconsistencies which may be arisen by integrating consistent knowledge bases.     

二阶分块PCA的人脸特征提取方法研究

为了提取更为有效的鉴别特征,在已有的二阶特征脸方法和分块主成分分析(PCA)方法上,提出了二阶分块PCA人脸特征提取方法.该方法对原始人脸图像和经重建得到的剩余图像分别运用分块PCA,将提取的一阶和二阶特征线性组合为一个特征矩阵,再进行分类识别.此特征能更充分反映人脸图像的低频和高频特性.采用ORL人脸库和FERET人脸库的实验结果表明该二阶分块PCA正确识别率优于普通分块PCA算法,具有较强的特征提取能力.     

基于FPGA的Montgomery模乘器的高效实现*

为了提高椭圆曲线密码处理器的模乘速度,本文提出了一种更有效且更适合硬件实现的Montgomery算法。改进的算法分析了基于CSA加法器的Montgomery模乘算法,提出了多步CSA加法器的Montgomery算法,该算法能够在一个时钟内做多次CSA迭代运算,可以有效地降低时钟个数,进而提高模乘速度。通过Modelsim仿真工具仿真,正确完成一次256bits Montgomery模乘运算只需要16个时钟周期。在Altera EP3SL200F1517C2 FPGA中的运行结果表明：71.5MHz的时钟频率下,完成一次256位的模乘运算仅需要0.22微秒。     

分割式Montgomery模乘运算的线性高基心动阵列新结构

本文基于提高并行性、加速模乘的思想，利用分割操作数的方法，提出了分割式Montgomery模乘算法（PMMM），并且基于C．D.Walter发明的心动阵列结构，提出了新的线性高基心动阵列模乘结构，较好地实现了PMMM。对于基r（r=2^w）的n位模乘运算，Walter使用（n＋1）（n＋2）个PF来实现Montgomery模乘，我们用n＋2个PE实现Montgomery模乘，最大并行性为Walter的2倍。将此结构应用于模幂运算，仅需一次预计算便可使得非平方模乘的输入输出延迟为walter中的1/2，且平方模乘延迟与其相当，从而提高了模幂的运算速度。当然，考虑到对速度和硬件资源的不同需求，我们也给出了使用n/2＋1个PE来计算模乘、模幂的实现算法，并做出了相应的数据分析。     

模块神经元网络中耦合时滞诱导的簇同步转迁

利用Courbage-Nekorkin-Vdovin神经元构建含有耦合时滞的模块神经元网络模型,通过数值模拟研究了耦合强度及耦合时滞对模块神经元网络簇同步放电特性的影响.研究结果表明,适当大的耦合强度可以诱导模块神经元网络达到簇同步.同时,研究发现耦合时滞可以诱导模块神经元网络出现簇同步转迁,且当时滞大小约为网络中所有神经元平均振荡周期的整数倍数时,模块神经元网络的簇同步现象能够间歇性出现.此外,研究结果表明时滞诱导的簇同步转迁对子网络内的耦合强度、子网络间的连接概率具有鲁棒性.     

有限域模乘专用指令设计

针对椭圆曲线密码算法中有限域模乘运算的需求,提出其专用模乘指令。利用指令域中的组参数实现算法多组模乘运算,通过对参数进行配置,使指令支持运算长度拓展,在模乘运算单元中实现Montgomery模乘算法,并设计素域和二进制域统一的硬件流水线,以及双域乘法器单元结构。实验结果表明,该有限域模乘指令和硬件运算单元具有较高的执行效率和较好的灵活性。