一类Boltzmann & Darwin混合寻优策略的收敛特性研究

基于文［１］提出的求解组合最优化问题的Ｂｏｌｔｚｍａｎｎ＆Ｄａｒｗｉｎ混合寻优策略，本文进一步分析研究了Ｂｏｌｔｚｍａｎｎ＆Ｄａｒｗｉｎ混合寻优策略非时齐情形下的收敛特性，包括全局收敛性和收敛速度。结果表明，Ｂｏｌｔｚｍａｎｎ＆Ｄａｒｗｉｎ混合寻优策略的收敛特性优于模拟退火（ｓｉｍ－ｕｌａｔｅｄ ａｎｎｅａｌｉｎｇ）方法的收敛特性。     

输入——输出迭代算法中松弛因子的确定

本文利用变分的手段提示了输入—输出迭代算法中松弛因子的取值规律。     

混凝土结构非线性全过程分析中的迭代控制法研究

在深入研究各类迭代控制法的基础上，针对混凝土结构的非线性全过程分析，在常规的位移和弧长控制法中引入了简单适用的考虑路径搜寻和单因加速的方法，明显改善了收敛情况。     

Iterative finite element model of nonlinear viscoplastic analyses for blended granular porous media

The iterative finite element model, in which an element is used to represent a single particle, is generated to analyze the global behavior of multiple-material aggregates of materially nonlinear viscoplastic particles. The generalized Maxwell model is used to define four types of specific nonlinear viscoplastic materials, which are the elasto visco-plastic matter with linear viscosity, the plasto visco-plastic material with linear viscosity, the elasto visco-plastic media with nonlinear viscosity, and the plasto visco-plastic media with nonlinear viscosity. The theory and relevant penalty iterative algorithm are developed to analyze the four representative mixed granular systems consisting of materially nonlinear viscous particles. To verify precision of stress calculation, solutions of an axis-symmetric radial flow problem are compared with results of the literature and they are in a great agreement. The results present here provide significant insight into the fundamental behavior of granular media under compaction conditions, including prediction of the overall aggregates stress-strain response.     

Turbo码概率-代数联合解码算法

Turbo码采用修正的BAHL et al.算法实现解码.这是一种基于软值的概率迭代解码算法.本文在保持Turbo码迭代软解码算法优点的基础上,充分利用Turbo码编码器结构这一确知条件,结合代数解码原理,提出了一种Turbo码概率-代数联合解码算法.该算法结合了概率解码和代数解码的优点,又有效避免了误差传播的发生,使Turbo码的纠错性能在原经典算法的基础上得到进一步的提高.该算法不仅为降低Turbo码的比特误码率和误差地板值提供了一种新的研究途径,而且因其更好的纠错性能而具有十分明显的实用价值.仿真实验结果显示,在比特误码率(BER)为10^-3~10^-4时,与经典Turbo码解码算法相比,采用该算法能获得0.1dB左右的编码增益.     

PERFORMANCE OF THREE STAGE TURBO－EQUALIZATION－DECODING

An increasing demand for high data rate transmission and protection over bandlimited channels with severe inter-symbol interference has resulted in a flurry of activity to improve channel equalization. In conjunction with equalization, channel coding-decoding can be employed to improve system performance. In this letter, the performance of the three stageturbo-equalization-decoding employing log maximum a posteriori probability is experimentally evaluated by a fading simulator. The BER is evaluated using various information sequence and interleaver sizes taking into account that the communication medium is a noisy inter symbol interference channel.     

线性分组码的神经网络译码

介绍了神经网络与线性分组码之间的关系，并在文献[1]的基础上证明了软判决译码与求解能量函数最大值之间的等价性，然后以(7，4，3)汉明码为例介绍了神经网络在循环码硬判决、软判决译码中的应用。     

DSP-based digital FM demodulation for GMSK signals

Demodulation of Gaussian Minimum Shift Keying (GMSK) using a limiter-discriminator is a low complexity alternative to coherent demodulation. This so-called digital FM demodulation is followed by clock recovery, sampling, and thresholding. Conventionally, clock recovery is done in hardware, and matched filtering is usually not possible when the Gaussian pulse is wider than a bit duration. We propose a clock recovery technique based on discrete-time processing of the demodulated baseband signal. This technique couples very nicely with a new maximum likelihood sequence estimator for the data that uses a whitening filter followed by a Viterbi decoder. The entire detection algorithm can be implemented in an efficient manner on a Digital Signal Processor (DSP). Computer simulation results are presented to show that the new algorithm performs better than the conventional slicer by as much as 5.5 dB.     

密码学函数迭代原理信息论分析

在密码编码学中,经常利用密码学函数迭代技术来实现密码算法,其所依赖的理论基础包括相关免疫理论,扩散准则,雪崩原理等.本文利用信息论原理以及随机过程理论给出了密码学函数迭代原理分析,得到了经过密码学函数迭代之后,输出为均匀分布时,输入输出互信息极限为零的充分必要条件,以及在一定条件下输入输出互信息收敛速度的一个上界.     

一种数字解码卫星电视接收机的原理分析

分析了一种数字解码卫星电视接收机，该机能接收利用ＭＰＥＧ－２进行数据压缩，采用ＳＣＰＣ或ＭＣＰＣ方式上星的卫星电视信号，具有选单显示，断电现场保护等功能。并对该机进行了原理分析，对主要集成电路的基本原理也作了简介，并给出了软件的部分主要流程及功能分析。