lab4

实现拼图游戏功能，锻炼智力，供学习开发，了解另一种开发工具！(The realization of jigsaw puzzle function, mental exercise for the study and development, to understand another development tool!)

使用EM算法，估计8PSK调制方式下的信噪比(EM algorithm using the estimated signal to noise ratio under 8PSK modulation)

matlab平台上dsp代码自动生成，应用matlab和ccs的联合仿真提高开发周期(Matlab platform dsp automatic code generation, Application ccs and Matlab simulation to improve the joint development cycle)

file for generating inter-arrival and service time in a banking system generation of uniformly distributed random numbers

gps信号的自相关，以及boc信号的自相关(gps signal autocorrelation, and boc autocorrelation)

不仅实现了ASK的调制信号，而且实现了加噪后的调制信号的解调过程。(Not only to achieve the ASK modulation signal, and the realization of the noise signal after the demodulation process of modulation.)

核主成分分析和支持向量机方法相结合，用于数据分类和预测。(Kernel principal component analysis and support vector machine method combined for data classification and prediction.)

一个关于qam调制方法的小程序，是用matlab语言编写的(Qam modulation method on a small program is written in matlab)

包括各种频域滤波器：经典频域理想低通滤波器 ，经典频域巴特沃斯低通滤波器 ，经典频域指数低通滤波器， 经典频域理想高通滤波器 ， 经典频域巴特沃斯高通滤波器， 经典频域指数高通滤波器,经典频域带通滤波器 ,经典频域带阻滤波器(Including a variety of frequency domain filter: Classic ideal low-pass filter frequency domain, frequency domain classic Butterworth low-pass filter, the classic low-pass filter frequency domain index, the classic ideal high-pass filter frequency domain, frequency domain Butterworth classic Andean high-pass filter, the classic high-pass filter frequency domain index, the classical frequency domain band-pass filter, band stop filter frequency domain classics)

In this paper, we show how support vector machine (SVM) can be employed as a powerful tool for $k$-nearest neighbor (kNN) classifier. A novel multi-class dimensionality reduction approach, Discriminant Analysis via Support Vectors (SVDA), is introduced by using the SVM. The kernel mapping idea is used to derive the non-linear version, Kernel Discriminant via Support Vectors (SVKD). In SVDA, only support vectors are involved to obtain the transformation matrix. Thus, the computational complexity can be greatly reduced for kernel based feature extraction. Experiments carried out on several standard databases show a clear improvement on LDA-based recognition