一种自动布线方法：多级线探索法

本文介绍了一种在微机上实现的印制板自动布线方法──多级线探索法及其编程方法。这种方法是基于完备布线算法的概念，通过分析李氏算法和传统线探索法的不足之处而得到的。虽然它也是一种线探索法，但有较传统线探索法更强的探索能力，不需要结合李氏算法而能单独完成布线任务，是一种较理想的算法，通过实验获得了较为满意的效果。     

SnO2超微粒子薄膜的气敏特性研究

作者用自行设计的直流气体放电活化反应蒸发装置制备出平均粒径约为40nm的SnO_2超微粒子薄膜.研究了不同氧分压下所得SnO_2超微粒膜的形貌、结构和组成等特性,以及不同样品对各种易燃气体的气敏特性,得出了灵敏度随氧分压及灵敏度随工作温度的变化曲线.     

应用粒子动态分析仪研究水力旋流器流场

利用粒子动态分析仪(PDA),对水力旋流器内固相颗粒流场做了全面的测定,得出水力旋流器内固相颗粒的流态、浓度及粒度分布,为进一步查清水力旋流器的工作机理提供了重要理论依据。     

Sequential Decoding of Convolutional Codes for Rayleigh Fading Channels

The performance of sequential decoding of long constraint length convolutional codes is evaluated for Rayleigh fading channels. Sequential decoding is not practical below a certain theoretical signal-to-noise ratio, and these theoretical limits are calculated for a number of modulation methods and code rates. As an example, with BPSK modulation, soft decisions and code rate 1/2, the theoretical signal-to-noise ratio per information bit is 5.7 dB. Above this limit the bit error rate can be made arbitrarily small by increasing the constraint length at no significant complexity cost. Furthermore, it is shown that with carefully chosen quantization steps, 8 level uniform quantization gives a negligible loss also for sequential decoding on a Rayleigh fading channel. Simulation results using 8 level quantization correspond well with the theoretical performance bounds. Also, the performance on a correlated channel with finite interleaving has been obtained. With an interleaver depth of 50×50 and a normalized doppler frequency equal to 0.01 we are only 0.5 dB away from the performance with perfect interleaving. Finally, bit error rate results show this scheme to compete well with Turbo codes.     

SHAPE GENERALIZED ISOTHERMAL EFFECTIVENESS FACTOR FOR FIRST-ORDER KINETICS

A generalized effectiveness factor equation (Eq. (32)), in terms of modified Bessel functions, is derived for a catalyst pellet of arbitrary shape. The derivation is based on utilizing an appropriate one-dimensional approximation for the Laplacian in an arbitrary shaped body subjected to a uniform external concentration field. The comparison of the result with the available expressions for various geometries is highly satisfactory. It unifies the expressions for the three fundamental shapes, viz., infinite slab, infinite cylinder and sphere, and also compares very well with the exact solutions for finite shapes over the entire range of the Thiele modulus.     

溶胶-凝胶法制备HAP骨水泥材料的研究

文章对用溶胶－凝胶法（简称SOL－GEL）烧结制备HAP骨水泥材料的工艺条件进行了研究，并探索了HAP粉剂与柠檬酸水溶液构成和凝结体系的可行性，结果显示出该材料作为医用骨水泥材料的可能性。     

利用框架理论对信号进行重建

在Hilbert空间中讨论了框架的一些重要的性质,定义了一个框架算子,并讨论了该算子的性质。根据实际问题构造了一框架,使信号在此框架下的投影正好为非均匀采样值。给出了一种框架界的估计方法和重建原信号的迭代算法,最后,用仿真验证了此算法的有效性。     

基于DOCSIS标准的HFC数据加密算法的分析

基于DOCSIS标准的增强型基线保密性接口（BPI＋）采用56bit DES加密分组数据，采用二密钥三重DES加密通信量加密密钥。这种多重加密技术的应用为HFC网络的数据传输提供了相当于或更好于专线级的网络安全与加密性能。     

一种有效的优化的多载波通信系统的功率分配算法

本文提出一种关于DMT系统传输纯数据流业务的最佳的功率分配算法，该算法使用了一种有效的表格查手工艺工和拉格朗日乘法器对分搜索办法，能够较快的收敛到最佳的功率点。同时，易于用硬件和软件实现。     

基于局部多项式逼近的变参数梯度估计算法

本文分析了梯度辨识算法跟踪时变系统的缺点，提出了一种新的基于局部多项式逼近的变参数梯度估计算法。本算法具有良好跟踪性能特别适用于时变系统跟踪且其对噪声的敏感程度与一般梯度估计方法相当。仿真结果验证了其良好性能。