在算法上

In this paper, firstly, some basic signal process are discussed, such as signal fil...

论文中首先对语音信号的基本处理问题进行了分析和对比，然后在自己设计的基于TMS320VC5402的DSP实际系统上，进行了语音处理过程的滤波、采样、傅立叶变换和谱包络提取的算法实现研究，讨论了在算法的DSP实现方法，分析了运行实验结果。

The adopting of the recursion type computation avoids the floating point calculation and has accelerated the operating.And the adopting of the symmetrical analysis method has simplified the Bresenham programme code designing of the straight line on any direction.

在研究实践中发现，对于任意方向的直线Bresenham完整生成算法和程序设计，尚有必要作进一步分析和研究。1问题描述Bresenham直线生成算法的基本思想：在屏幕上给出直线的起点和终点坐标，由起点出发，通过若干次走步达到终点。

In the proposed algorithm, we make use of the Lipschitz condition to determine lower bounds of functions over each rectangle.

在算法中，我们运用Lipschitz条件来确定目标函数和约束函数的在每个n-矩形上的上下界，对于n-矩形的分割，我们采用选择珩矩形最长边的二分法，同时我们采用了一些矩形删除技术，在不大幅增加计算量的前提下，起到了加速算法收敛的效果。

Second, a nonuniform bilateral filtering algorithm for point set surface and surface attributes is developed, in which the tangent shift of points is avoided. The algorithm can be applied when points are sampled nonuniformly.

其次提出了基于双边滤波对曲面和定义在曲面上的函数的非一致双边滤波算法，该算法不仅避免了滤波过程中点的切向漂移，而且考虑了点采样的密度使得算法可以适应非均匀采样。

It inherits merits of trondional AdaBoost algorithm, such as the high rate, speed of recognition, and etc. In order to solve the problem of overfit, which the classical AdaBoost algorithm brings, a weight-adjusting method based on the distribution of samples is brought forward.

它不仅继承了传统AdaBoost算法的识别率高、识别速度快等优点，针对AdaBoost算法的退化现象，根据样本在权重上的分布情况，对权重进行适当的调整，解决了经典AdaBoost算法中存在的退化现象。

Use SNMP agreement and join integratedly to decide theorematic, put forward develop of link layer network to attack discovery is algorithmic, undertook introductory to this algorithm, pass this algorithm, can go up to attack to the develop of current network in stand-alone the circumstance undertakes discovering, the network equipment that in the light of information substantial level differs uses different strategy, achieve the result of case of connection of network of real report physics.

综合使用SNMP协议和连接判定定理，提出链路层网络拓扑发现算法，并对该算法进行了介绍，通过该算法，可以在单机上对当前网络的拓扑情况进行发现，针对信息丰富程度不同的网络设备采用不同的策略，达到真实反映物理网络连接情况的效果。

The unfeasibility of discrete event driven simulation for high-speed network is discussed; two simulation algorithm based on time driven simulation are given, one of which is as precise as that based on discrete event simulation, but is faster, and the other gets less precision, but is much faster than the former. Comparing with the algorithm based on time driven fluid simulation, both of the proposed algorithms support the succeeding simulation for high-level protocol.

其中一种算法在精度上与离散事件驱动仿真算法一致，但仿真速度明显加快，另一种算法存在一定的仿真误差，但仿真速度要高得多；与时间驱动的流体仿真相比，这两种方法都是针对离散系统的仿真，能提供对上层协议仿真的支持。

Experimental results show in detail that the proposed algorithm overweighs other corresponding algorithms not only in speed but also in precision due to its fast convergence.

实验结果表明，由于算法的快速收敛性，使其在速度上和精度上均优于其它相应算法。

This paper aims at the time performance of the entire system and uses the task flow graph as the system model.Based on it,an evaluation function based on the priority is achieved,then a new algorithm combining the search space smoothing technology with discrete particle swarm is proposed and whose parameters can be adjusted automatically.At last,the combination problem is addressed.

以提高系统时间性能为目标，利用任务流图构造系统模型，在其上实现了基于优先权的评价函数，提出了搜索空间平滑技术与离散粒子群算法相结合的软硬件划分算法，并且解决了两者的融合问题，并能根据系统信息动态适应调整算法参数。

Parameters" interpolation, adaptive spectrum enhancement and pulse dispersion also are adopted to improve the continuity. The synthetic speech of MELP vocoders sound much more natural and perceivable than the traditional vocoders".

本文以美国联邦标准2.4kbps－MELP算法为基础，在MATLAB上建立起了分析MELP算法的软件平台，对其性能进行了分析并提出了一些改进的建议；另外还针对MELP算法的特点对其软硬件实现进行了探讨。

I didn't watch TV last night, because it .

昨晚我没有看电视，因为电视机坏了。

Since this year, in a lot of villages of Beijing, TV of elevator liquid crystal was removed.

今年以来，在北京的很多小区里，电梯液晶电视被撤了下来。