抵御已知/选择明文攻击的混沌图像加密算法

论文在对现有一类典型图像混沌加密算法的分析基础上,提出了一种改进的图像混沌加密算法。该算法引入小波变换,可以有效地克服一些混沌加密算法不能抵御已知/选择明文攻击的缺陷。     

基于非均匀星座的改进ARQ方案及其性能分析

本文针对调制阶数大于4的MPSK和MQAM调制，在重传中采用非均匀星座图和符号比特重新排序，提出了一种改进的ARQ方案。通过对AWGN信道下ARQ方案进行理论分析和数值仿真，表明基于非均匀星座的新方案在信道条件较差时能有效地提高重要比特的可靠性，若重传中结合符号比特重排和分组合并，则可使接收端解调合并后的比特可靠性趋于均匀且总体得到提高，从而有效地减少重传次数，提高系统的吞吐率。由于本文所提方案并不改变调制解调规则和数据分组长度，故容易实现和控制。     

基于小波基优化和动态比特分配的数字音频压缩编码方案

提出了一种新颖的基于自适应小波基优化选择和心理声学模型相结合的数字音频信号的透明质量编码方法，保证固定失真水平上使每帧信号的变换系数的动态分配的比特数最少，并且利用动态码本的方法来消除音频信号的统计冗余，进一步压缩比特率，对于抽样率为４４．１ｋＨｚ每样值用１６比特线性码表示的光盘单声道音乐信号可以压缩到６４ｋＢＰＳ左右。     

基于余弦-指数混沌映射的分块图像加密算法

为平衡混沌映射中结构与性能的关系,保证加密系统安全性,提出一种基于余弦-指数混沌映射的分块图像加密算法。首先,通过非线性指数项对引入了Tent种子映射的余弦映射进行调制,构造新型余弦-指数混沌映射,并利用SHA-256函数产生与明文相关的密钥,生成随机性较强的混沌序列,实现一次一密;然后,基于拉丁方和位级转换,通过两轮拉丁方索引和比特位拼接,分别设计双重拉丁方和扩展比特位算法,并结合二维约瑟夫序列,对块间预置乱后的明文进行块内置乱,实现不同分块的差异化置乱;最后,基于Zig-Zag变换,采用环状仿Zig-Zag变换设计交叉Zig-Zag变换方法,将中间密文与混沌序列进行双向非线性扩散,实现同时改变像素位置与大小,完成图像加密。实验结果表明,该算法密钥空间大,能有效抵御差分分析和统计分析等典型攻击,具有较好的加密效果。     

一种基于改进Logitic映射的跳频序列生成算法

在扩频序列生成算法中,针对周期抽取算法平衡性和比特重排算法稳定性的欠缺,基于改进的Logitic映射,结合采用比特抽取和扰动的方法,提出了一种改进的算法来构造跳频序列, 提出的算法不需要考虑与其他映射或者序列进行同步的问题,很大程度上缩减了迭代次数。仿真结果表明,提出的算法可以提供相关性更加优良的序列,且均匀性、平衡性上优于周期抽取比特重排算法,汉明相关性与理论参考值近似重合。     

功率裸芯片的测试与老化筛选技术

由于存在高温度、大电流等问题,传统的测试与老化筛选功率管的方法不能完全适用于功率裸芯片.介绍了采用临时封装夹具来测试与老化筛选功率裸芯片的技术,并根据功率裸芯片的实际情况,阐述了测试与筛选功率裸芯片的方法.     

Packet switching start-stop bits generation based on bifurcation behavior of light in a micro ring resonator

This paper proposes the interesting nonlinear behavior of light known as bifurcation, where the use of such behavior in a micro ring resonator to form the secure digital codes for optical packet switching application is demonstrated. A new concept of the stop-start bits in an optical packet switching protocol is formed by using the bifurcation codes. Bifurcation is introduced when light is input into a nonlinear micro ring device, where the refractive index of an InGaAsP/InP is one of device parameters. The other parameters of the device are coupling coefficient (K) and the ring radius (R), where the ring radii used are ranged from 5 to 10 μm. Simulation results obtained have shown that the packet switching data can be secured by using the generated start-stop bits as the secured codes.     

On isoperimetric problems for domains with partly known boundaries

In the present paper, we discuss the isoperimetric problems for domains with partly known boundaries, i.e., the problem of determining a domain that minimizes the capacity functional in the class of plain double-connected domains having the same fixed area and outer boundary. The formulas for capacity variations obtained in this paper allows us to formulate necessary conditions.It is proved that the convexity of the fixed outer boundary implies the convexity of the inner boundary corresponding to an optimal domain. Then, we discuss the case where the fixed part of the boundary is a square.Further, we consider similar problems with more complicated functionals. We introduce the concept of a minimal function in the class of equimeasurable functions. This concept allows us to unify the approach to all of these problems. At the end, we produce a hypothesis that, if proved, would enable us to characterize the shape of the optimal domains in the isoperimetric problems mentioned above.The author wishes to express his appreciation to Dr. K. A. Lurie for his help and unceasing attention.     

关于不确定度的教学探讨

文章澄清了大学物理实验中关于不确定度内容的教学中遇到的一些困惑,给出了相应的解决办法.