基于广义秘密共享方案的可生存存储系统研究

高生存性是分布式存储系统设计所追求的重要目标之一。文中提出了一种改进的广义秘密共享方案,给出了它的形式化定义及实现方法;在此基础上设计了一个可生存存储系统模型,讨论了它的体系结构与工作流程;建立起了机密性与可用性评价体系,对该系统的机密性与可用性进行了详细评价。该系统针对实际应用中存储服务器所存在的逻辑分组与特权分级的情况,采用了广义秘密共享方案来实现信息的存储与重建,可以达到较高的机密性和可用性,具有较强的实用性,为构建高生存性存储系统提供了可能。     

TD-LTE系统小区搜索定时同步的改进算法

通过对TD-LTE系统小区搜索定时同步的研究,结合Zadoff-Chu序列自身的特性,给出了一种基于SSS的定时同步算法,利用SSS信号的时域共轭对称性正确地估计定时同步点。理论分析和仿真结果表明,该算法性能良好,复杂度低,在存在频偏的情况下也能够准确地估计SSS定时同步点的位置,较常规算法可靠性有了很大提高。     

半透明材质在Mental Ray中的实现

基于Henrik Wann Jensen的SSS数学模型Single Scattering，在Mental Ray下编写shader实现半透明效果，并在实现中对其运算速度、可控性做分析和调整。最终得到玉石质地模拟的良好效果，真实感极强。     

一种基于DCT的改进D-LDA人脸识别算法

D-LDA法是一种简单有效的线性特征提取方法,但在实际应用中往往存在以下两个问题：（1）去除Sb零空间的同时往往间接丢失了Sw零空间中的有用信息;（2）优化准则函数并不直接与识别率相关。离散余弦变换（DCT）能够有效地对原始图像的信息进行压缩,提出一种DCT与改进的D-LDA相结合的方法,首先利用DCT降维,然后在低维空间中应用一种改进的D-LDA方法进行特征提取,最大限度地克服D-LDA的不足。实验结果证明这种方法能获得较高的识别率。     

基于规范化KDDA的人脸识别算法

传统的PCA和LDA算法受限于“小样本问题”,且对像素的高阶相关性不敏感。论文将核函数方法与规范化LDA相结合,将原图像空间通过非线性映射变换到高维特征空间,并借助于“核技巧”在新的空间中应用鉴别分析方法。通过对ORL人脸库的大量实验表明,该方法在特征提取方面优于PCA,KPCA,LDA等其他方法,在简化分类器的同时,也可以获得高识别率。     

NAVAL ENIGMA: SEAHORSE AND OTHER KRIEGSMARINE CIPHER BLUNDERS

Describes major weaknesses in a naval Enigma cipher codenamed Seahorse, and how it and similar naval Enigma ciphers were solved by US Navy codebreakers during WWII.     

14型高分辨率彩色显示管的性能及其应用

本文介绍了ＢＭＣＣ生产的１４型高分辨率彩色显示管的性能。简单的讨论了表面处理、点距、电子枪、偏转线圈、管子在显示器中的应用，以及如何满足ＭＰＲ一刀标准。并给出了管子及其组装产品（ＩＴＣ产品）的性能规格。     

一种TD-LTE系统下行辅同步信号检测方法

基于相干检测和使用部分相关的非相干检测,提出了一种适用于TD-LTE系统下行辅同步信号SSS(Secondary Synchronization Signal)的检测方法。根据SSS索引号(m0,m1)对应关系,通过优化降低了索引号m1的检测运算量。仿真结果表明,在不同的信道环境下,本文提出的SSS检测方法能较好地工作,具有稳定、检测复杂度低等优点。     

零空间边界Fisher分析法及其在人脸识别中的应用

边界Fisher分析（MFA）是一种有效的特征抽取方法,但在人脸识别的应用中会遭遇小样本问题。基于此,提出一种利用零空间法求解MFA优化准则的算法。该算法通过在MFA的类内散度矩阵的零空间中最大化MFA类问离散度得到最优投影向量,从而避免MFA方法所遇到的小样本问题,同时也保留了包含在类内散度矩阵零空间中的鉴别信息。在标准人脸库上的识别实验结果表明,该算法的识别率高于LDA和MFA,并且较容易选择其最优低维特征空间的维数。     

Sea surface salinity retrievals from HUT-2D L-band radiometric measurements

The interest of the scientific community in global climate has been constantly increasing in the last years. Much effort has been devoted to better understand the water cycle and its role in global climate regulation. This is one of the objectives of the European Space Agency (ESA) Soil Moisture and Ocean Salinity (SMOS) mission, approved in May 1999 and successfully launched in November 2nd 2009. Collecting brightness temperature measurements by means of a new type of instrument, a synthetic aperture radiometer operating at L-Band (MIRAS: Microwave Imaging Radiometer by Aperture Synthesis), the SMOS mission will provide regular and global maps of sea surface salinity (SSS) and soil moisture (SM), two key parameters to describe the water cycle. Synthetic aperture brightness temperatures have been simulated since the early 90s, during the MIRAS technological studies, and both SSS and SM retrievals have been tested with simulated data first, and later with real data to assess and improve the instrument performance. Nevertheless, except for the processing of the data collected by the SMOSillo (MIRAS Demonstrator) on June 20th, 2006, Duffo et al. (2007) and Camps et al. (2008b), these outcomes have never been validated using sea surface brightness temperatures coming from a real synthetic aperture radiometer. The aim of this study is thus to test some of the techniques proposed in previous years to improve the SSS retrieval, in particular only Level 2 techniques (i.e. converting calibrated and geo-located brightness temperature into SSS maps) will be investigated. To do so, brightness temperatures resulting from the SMOS Salinity Demonstrator Campaign held in August 2007 will be used. In that campaign the Helsinki University of Technology-2 Dimensional (HUT-2D) radiometer flown over both over land and sea. The part of the campaign conducted over the sea consisted of two series of flights over a very fresh water plume characterized by a strong SSS gradient (from 0 to 4 psu) in the Gulf of Finland. In-situ auxiliary data was collected simultaneously with the radiometer measurements. The positive outcomes in these very challenging conditions (due to the very low brightness temperature sensitivity to SSS, the lack of accurate models of the sea water dielectric constant at low SSS (Klein & Swift (1977)), and the weak radiometric sensitivity of the HUT-2D radiometer) demonstrate the importance of data pre- and post-processing to improve the results. Removing both brightness temperature biases and salinity retrieval biases and further averaging of the results, permits estimating SSS with an rms error on the order of 1 psu, which is comparable to the SMOS Level 2 expected accuracy.