MATLAB_commend

基于卡尔曼滤波方法的目标跟踪函数的一种机动目标的预测法(Based on kalman filtering method of target tracking function of a kind of maneuvering target forecast method )

在matlab环境下，编程实现用拉普拉斯算子改善图像的反差增强程序(In the matlab environment, programming with Laplacian to improve image contrast enhancement procedure)

some functions for face recognition

旋转点目标的SAR成像算法。分析了旋转对常规成像的影响，之后进行了补偿。(Rotating point target of SAR imaging algorithm. Analysis of the impact of rotation on conventional imaging, after the compensation.)

file is containing the exponential distribution and random numbers

基于端射阵的平面阵列方向图研究，对端射阵优化，利用切比雪夫加权。(Figure based on end-fire array of planar array direction of end-fire array optimization, the use of Chebyshev weights.)

该代码实现了在鼠标框定区域内的Harris Corner的检测(This code deals with the detecting of harris corners with matlab)

covolution code in matlab

说明：  第 5 代移动通信系统 (5G) 是面向 2020 年之后的新一代移动通信系统 , 其技术发展尚处于探 索阶段 . 结合国内外移动通信发展的最新趋势 , 本文对 5G 移动通信发展的基本需求、技术特点与可 能发展途径进行了展望 , 并分无线传输和无线网络两个部分 , 重点论述了富有发展前景的 7 项 5G 移 动通信关键技术 , 包括大规模天线阵列、基于滤波器组的多载波技术、全双工复用、超密集网络、自 组织网络、软件定义网络及内容分发网络 . 本文还概括性地介绍了国内 5G 移动通信的相关研发活 动及其近期发展目标 .(The fifth generation mobile communication system (5G) is a new generation of mobile communication system after 2020. Its technological development is still in the exploratory stage. Combining with the latest trends of mobile communication development at home and abroad, this paper prospects the basic needs, technical characteristics and possible development ways of 5G mobile communication development, and divides wireless transmission and wireless network into two parts, focusing on rich. Seven key technologies of 5G mobile communication, including large-scale antenna array, filter bank-based multi-carrier technology, full duplex multiplexing, ultra-dense network, self-organizing network, software definition network and content distribution network, are prospected. The related research and development activities and recent development goals of 5G mobile communication in China are also briefly introduced in this paper.)

传统的决策树能较好地解决分类问题，但当类的个数增多时，所产生的单一决策树就会变得复杂，同时概括能力降低。 该文采用基于层次分解的方法通过产生多层决策树来处理多类问题。 与传统的单一决策树比较， 基于层次分解的决策树在处理多类问题时有许多的优势。(induction learning,decision tree,id3 ,hiberarchy decomposition)