独立成分分析的研究进展

介绍了独立成分分析（ICA）的基本模型及其假设、含混性、非高斯性度量和通用求解过程。讨讨了目前ICA的几个研究方向的发展现状和面临的问题，分析了ICA基本模型和几种扩展模型的求解算法，包括盲反卷积、卷积混和的盲分离、非线性瞬时混合的盲分离，提出了ICA未来理论和应用研究中的开放课题．     

广义椭圆曲线数字签名动态用户认证机制

基于口令的身份认证机制是分布式系统安全体系的重要组成部分。利用一个无需求逆的广义椭圆曲线签名算法构造了一个动态远程用户认证方案:GECSA方案。该方案基于“质询响应”机制,无需时戳,其安全性基于单向Hash函数和椭圆曲线离散对数计算的困难性,包括注册阶段、登录阶段、认证阶段和更改口令阶段,允许用户自主选择并更改口令,实现了双向认证;具备完备性和不可转让性,是一个低开销的、安全的远程用户认证机制。     

Merkle Tree Digital Signature and Trusted Computing Platform

0 IntroductionDigital signatureis a key technique to provide datainteg-rity andsource authenticationcapabilities to enhance da-ta trustworthiness in trusted computing platform. Most cur-rent digital signature schemes are based on complex mathe-matical problems , such as integer factoring, discrete log-arithm,and discrete logarithm on elliptic curves . The hard-ness of these underlying mathematical problems is really un-certain at present . Merkle treeis a method used for authenti-cation and di…     

Threshold Subliminal Channel Based on Designated Verifier Signature

The subliminal channel is used to send a secret message to an authorized receiver; the message cannot he discovered by any unauthorized receivers. Designated verifier signature （DVS） provide authentication of a message, we design a DVS scheme with message recovery mechanism and use it as a subliminal channel. In order to share a message among n users securely and allows t or more users can reconstruct the secret in dynamic groups, we combine both subliminal channel and （t, n） threshold cryptography. Then we proposed a threshold subliminal channel which can convey a subliminal message to a group of users based on message-recovery designated verifier signatures. Reconstructing the subliminal message relies on the cooperation of t or more users in the group and they can verify the validity of the subliminal message. Security and performance analysis show that the proposed scheme is secure and efficient.     

A New Algorithm for On-Line Handwriting Signature Verification Based on Evolutionary Computation

The paper proposes an on-line signature verification algorithm, through which test sample and template signatures can be optimizedly matched, based on evolutionary computation （EC）. Firstly, the similarity of signature curve segment is defined, and shift and scale transforms are also introduced due to the randoness of on-line signature. Secondly, this paper puts forward signature verification matching algorithm after establishment of the mathematical model. Thirdly, the concrete realization of the algorithm based on EC is discussed as well. In addition, the influence of shift and scale on the matching result is fully considered in the algorithm. Finally, a computation example is given, and the matching results between the test sample curve and the template signature curve are analyzed in detail, The preliminary experiments reveal that the type of signature verification problem can be solved by EC.     

A New Digital Multilevel Proxy Signature Scheme Based on Elliptic Curve Cryptography

Based on the analysis of elliptic curve digital signature algorithm（ECDSA）,aiming at multilevel proxy signature in which the original signer delegates the digital signature authority to several proxies and its security demands, a new multilevel proxy signature scheme based on elliptic curve discrete logarithm problem （ECDLP） is presented and its security are proved.     

基于盲源分离的无线视频通信研究与仿真

提出了一种多路视频压缩信号在无线信道中的盲源分离解决方案.该方案选用IEEE 802.11b作为无线通信标准及XviD作为MPEG-4视频压缩方法,建立了一个MPEG-4无线视频传输系统及其仿真模型.该模型用盲源分离方法来解决多个视频信号的分离问题.仿真通过多次比较得出一种基于信号循环平稳性的NWCW-CS,其分离效果最佳及收敛速度最快.仿真结果还表明,该算法可以很好地解决视频信号的混叠问题,视频信号由XviD解压后的图像与原先一致,同时满足视频传输的实时性要求.     

基于小波自适应量化的水印算法

为了有效地保护数字媒体的版权,提出了一种基于小波的自适用量化的水印算法,使水印算法能确定水印的最优嵌入能量来量化小波系数,实现了水印不可感执行和鲁棒性之间的折衷,同时,水印的提取无须求助于原图像,很好地实现了水印的盲检测.实验结果和攻击测试表明,提出的算法对JPEG/JPEG2000压缩、添加噪声、剪切、亚抽样、低通滤波等多种攻击有较强的鲁棒性.     

基于盲数理论的最优化方法与程序实现

使用盲数表达优化设计中的不确定变量,结合常用的基于实数变量的优化算法,提出了基于盲数的优化方法.该方法从微观的角度分析和描述设计变量与优化参数之间的关系,给出优化问题的盲数解.盲数解不但给出了设计变量的取值,而且还给出了不同取值时优化对象处于最优状态的可靠性的评价.     

基于前馈复式的BSS去噪方法

在BSS中，含噪声信号的盲分离是困难的，因为存在太多的未知条件。针对噪声环境，提出了一种基于前馈复式BSS去噪新方法。由于BSS算法的目的是尽可能恢复源信号，因而其实际输出信号总是由某个较强的源信号迭加另一些较弱的源信号及噪声组成。该方法是把那些较弱的源信号及噪声又假定为一个新的混合源信号Snew，那么通过适当的选择和组合前一次盲分离的结果作为下一次盲分离的源，反复使用盲分离技术就可把这个假定的Snew分离出来，再将Snew与选定的某个恢复源信号取谱差，达到消除噪声，提高分离结果性能指标的目的。