guangfu

Matlab: simulink mdl files for some basic queues.

说明：  it is a part of a genetic algorithm applied to image segmentation

QPSK信号在加入白高斯噪声（AWGN）后，用matlab进行仿真，并绘制了误码率图形。( The QPSK signal after joining the white gaussian noise (AWGN), carries on the simulation with matlab, and has drawn up the error rate graph. )

在matlab中编制Levy法辨识程序并进行测试(Identification procedures for the preparation of the Levy method in matlab and tested)

旅行商遍历50个城市，最短线路的求法问题(Traversing 50 cities TSP, the shortest route problem Method)

Least mean squares algorithm for smart antenna

The (a) section of Figure 1 illustrates the Line Echo Canceller (LEC) residing in the access network where echoes occur. Other voice processing modules, such as Automatic Level Control (ALC) and Noise Reduction (NR), are shown merely for illustration purposes. On the terminal side, both acoustic and electrical echo can occur. If the voice terminal is a speaker telephone set with an open acoustical path from the speaker to the microphone, the effect of acoustical echo is combined with the effect of the network echo. Because there is only a miniscule delay between the electrical signal and its echo, the speaker/listener is exposed to an effect known as side tone. Typically, instead of mounting a LEC at the terminal, side tone circuitry is placed to control the amount of the returned energy. The (b) section of Figure 1 shows examples of echo canceller signals representing the following:

说明：  自动寻找阈值，实现图像分割。很好用啊！Matlab 编写！(Automatic find threshold, realize image segmentation. Very good! Matlab! )

depso clustering algorithm written in matlab

说明：  ：DBSCAN是一个基于密度的聚类算法。该算法将具有足够高密度的区域划分为簇，并可以在带有“噪声”的空间数 据库中发现任意形状的聚类。但DtLqCAN算法没有考虑非空间属性，且DBSCAN算法需扫描空间数据库中每个点的e一 邻域来寻找聚类，这使得DBSCAN算法的应用受到了一定的局限。文中提出了一种基于DBSCAN的算法，可以处理非空 间属性，同时又可以加快聚类的速度。(: DBSCAN is a density-based clustering algorithm. The algorithm has a sufficiently high density area is divided into clusters, and to be with the &quot noise&quot found in the spatial database clusters of arbitrary shape. But DtLqCAN algorithm did not consider non-spatial attributes, and spatial database DBSCAN algorithm to be scanned for each point e in the neighborhood to find a cluster, DBSCAN algorithm which makes the application subject to certain limitations. In this paper, an algorithm based on DBSCAN can handle the non-spatial attributes, can also speed up the clustering speed.)