modulation

研究表明超高斯分布更加贴近语音信号的实际分布，然而语音信号很难用单一的概率密度 函数准确描述，针对这一情况，提出了一种用超高斯混合模型对语音信号幅度谱建模的新方法，并推导了 基于此模型的幅度谱最小均方误差估的估计式。仿真结果表明：与传统的短时谱估计算法相比，该算法不 仅能够进一步提高增强语音的信噪比，而且可以有效减小增强语音的失真度，提高增强语音的主观感知 质量。 (Recent research indicates that the speech spectral amplitude distributions could be fairly described with super-Gaussian probability density function. However, the complexity of speech signal determines that the distribution statistics ofspeech signal could not be well described by single simple function. Thus a super-Gaussian mixture model for speech spectral amplitude is proposed, and with this model, a minimum mean-square error (MMSE) estimator for speech signals spectral amplitude is derived. The simulation results show that this algorithm based on Gaussian and super-Gaussian speech model could achieve better noise suppression and lower speech distortion as compared with the conventional short-time spectral amplitude estimation algorithm. )

文件中包含了Turbo码和卷积码误码率的对比，其中Turbo码采用了BCJR译码算法，迭代次数分别为1、2、3、4次。文件中包含了源代码和分析文档(File contains a comparison of Turbo code and convolution code BER, which Turbo code decoding algorithm used BCJR iterations respectively 1,2,3,4 times. File contains the source code and documentation analysis)

模式识别 关于GMM+EM+BIC，请运行附件的程序。主程序：VASceneAnalysisFH7_ICMEFull_2009.m (Pattern Recognition on the GMM+ EM+ BIC, please run the attached file. Main program: VASceneAnalysisFH7_ICMEFull_2009.m)

通过matlab代码，PLS部分最小二乘工具箱,可以广泛的应用于数据预测及数据分析。( By matlab code, PLS PLS toolbox, Can be widely used in data analysis and forecast data.)

simple code to find BER using convolutional codes

3D_mutlipeak with matlab

show an aliasing in sine wave by L. Braile

PSO 仿真软件包, 设计各种粒子群算法，最优化算法。(Particle ,Filtering,simulator)

对声音信号采样分析频谱,改变采样率，播放，音频信号加噪声(Spectrum analyzer of sound files)

此程序代码为基于遗传算法的PID整定，利用MATLAB编程实现。该程序通过遗传算法实现参数寻优，是一种不需要任何初始信息并可以寻求全局最优解的、高效的优化组合方法。(this procedure code based on genetic algorithms for PID tuning, using MATLAB programming. The procedures through genetic algorithm optimization of the parameters. is a not need any initial information and can find the global optimum, efficient optimization method.)