turbocodes

简单的Turbo码的编程，可以为Turbo码的研究提供参考(This is a simply program of Turbo code which can be used for some further researches.)

计算独立成分分析算法的matlab程序，通过串音误差来分析算法的收敛度。(The matlab program of independent component analysis algorithm is calculated, and the convergence of the algorithm is analyzed by crosstalk error.)

在高斯核函数的程序基础上加入了多项式核函数，构成了混合核函数，该混合核函数综合了两种单核函数的优点。(In the program on the basis of the Gaussian kernel function added polynomial kernel function, constitutes a hybrid kernel, the kernel mixed comprehensive advantages of two single-core functions.)

说明：  ofdm符号同步算法-schmidl&cox算法的matlab实现(OFDM Symbol Synchronization Algorithm-schmidl )

该程序是利用MATLAB来进行遗传算法和BP算法结合的好例子，可以为作参考(that the procedure was carried out using MATLAB genetic algorithm and BP algorithm combines a good example, we will be able to make reference to)

implémentation adaboost

经典的国外bitshop关于机器学习和模式识别教程，全是英文，内容详细(Classic foreign bitshop tutorial on machine learning and pattern recognition, all in English, detailed)

基于蒙特卡洛方法的主动声纳信号检测性能分析.主动声纳信号检测性能的分析上,目前在计算机仿真中一般假定混响包络的统计特性符合瑞利分布模型。基于此模型,已经有了较完善的理论。然而,在现代高分辨声纳系统中,混响包络的统计特性并不符合瑞利分布模型。此时在接收机工作特性分析时存在大量繁琐的公式推导。因此该文采用蒙特卡洛(M onte Carlo)统计试验方法,实现对瑞利分布混响背景下的主动声纳信号检测性能分析。结合对接收机工作特性曲线的仿真,得出了检测概率的理论值和仿真结果的误差曲线。误差曲线表明,蒙特卡洛方法在主动声纳信号检测的性能评估中是可行的。 更多还原 (Active sonar signal detection performance analysis, computer simulation is generally assumed that the statistical properties of the reverberation envelope distribution model of Rayleigh active sonar signal detection performance analysis based on the Monte Carlo method. Based on this model, has been a better theory. However, in the modern high-resolution sonar system, the statistical characteristics of the reverberation envelope does not comply with the Rayleigh distribution model. At this point in the receiver operating characteristic analysis, there is a lot of tedious formula derivation. In this paper, using the Monte Carlo (M onte Carlo) statistical test methods, active sonar signal detection performance analysis reverberation background Rayleigh distribution. The receiver operating characteristic curve simulation obtained the error curve of the theoretical value and the simulation results of the detection probability. The error curve shows that the Monte Carlo method in the assessm)

numerical computing matthematics matlab

随机森林算法在图像特征分类回归中的应用，通过结合神经网络进行更好的特征数据处理(Application of random forest algorithm in image classification and regression, better features by combining neural networks data processing)