ADC7787

matlab code (kMean.m) used kMean, clusturing. for Speech Recognition

陈杰编著的matlab宝典，高清的,可方便对matlab学习(matlab )

Zero Forcing Equalization for Complex Channel

matalb_各种命令集合（matlab使用中的各种命令的汇集，可以方便查找）(Matalb_ various command set)

《MATLAB6.0数学手册》 本教程按逻辑编排，自始至终用实例描述，既适用于初学者自学，也适用于高级 MATLAB 用户。可作为高等数学、数值分析、工程数学、数学建模、线性规划等课程的教 学参考书，也可作为科技工作者学习和使用MATLAB 的参考书，还可作为数学实验的教学 用书，特别适合用作理工科大学生学习数学课程的教学辅导书。(" MATLAB6.0 Mathematics Handbook," this tutorial logically arranged from beginning to end examples describing both self-study for beginners, but also to advanced MATLAB users. Can be used as advanced mathematics, numerical analysis, engineering, mathematics, mathematical modeling, linear programming courses teaching reference books, but also can be used as scientific and technological workers to learn and use MATLAB reference books, but also as an experimental teaching mathematics books, in particular, suitable for science and engineering students to study mathematics curriculum counseling books.)

使用基于峭度的FASTICA盲分离算法，实现对两个超高斯信号的分离，并从分离结果可总结出盲分离算法的两个不确定性。(FASTICA kurtosis-based blind source separation algorithm, to achieve the separation of the two super-Gaussian signals, and the separation results can be summarized from the blind separation algorithm of two uncertainty.)

这个是利用MATLAB实现的KNN算法的实现(This is achieved using MATLAB algorithm to achieve KNN)

In this project, the standard particles swarm optimization - named here (MyPSO) is simulated and tested by five well-known benchmark functions (Sphere, Ackley, Rastrigin, Rosenbrock and Schwefel p2.26). Then, a new mechanism- named here (Improved MyPSO) is introduced to improve the performance of the standard particles swarm optimization to achieve better results and avoid stagnation at bad results. This new mechanism is also tested by the same benchmark functions. The results of improved PSO are then compared with those of standard PSO and different published improved PSO (SPSO, PSO-XD, CPSO-S and PSO-P5).

模拟杂波的生成，包括高斯杂波，韦布尔杂波和K分布杂波，对数正态分布杂波(clutter simulation)

标准压缩感知（CS）理论决定了可靠的信号恢复是可能给M= O（KLOG（N / K））的测量。我们证明了它可以通过利用超越简单的稀疏性和可压缩性由包括价值观和信号系数的位置之间的依赖关系更加逼真信号模型大大降低Mwithout牺牲的鲁棒性。(The standard compressive sensing (CS) theory dictates that robust signal recovery is  possible from  M=O(Klog(N/K))  measurements. We demonstrate that it is possible to substantially  decrease  Mwithout sacrificing robustness by leveraging more  realistic signal models that go beyond simple sparsity and  compressibility by including dependencies between values and  locations of the signal coefficients.   )