approach-of-classification

针对混沌参数调制( C P M) 的电力线通信( P L C) 中谐波引起的窄带干扰， 两阶段动态估计方法根据最小 相空间体积( MP S V) 准则估计模型参数， 计算量大。为此， 提出将未知参数合并到增广状态矩阵的联合卡尔曼滤 波方法， 避免了专门的参数估计过程， 在提高增益性能的同时有效降低了计算量。方法的性能通过对混沌电力 线通信下的单音干扰和多音干扰的有效抑制得到了验证。 (For the chaotic parameter modulation (CPM) of the power line communication (PLC) in the narrow-band interference caused by harmonics, two-phase dynamic estimation methods with minimum phase space volume (MP SV) criterion estimate model parameters, calculation of volume. Toward this end, the unknown parameters into augmented state matrix of the joint Kalman filter, to avoid the special parameter estimation process, while improving the performance gain reduces the amount of calculation. Method of performance under the power line communication by chaotic multitone jamming tone interference and the effective suppression has been verified.)

............... Mobile robot tracking of pre-planned paths ..................................................................................................................

说明：  MATLAB7.X数字信号处理基础教程及源程序和课件。(MATLAB7.X digital signal processing based on source code and tutorials and courseware.)

deal with matlab source, matlab toolbox

主要是用matlab实现的数字水印，其中用到的技术是lsb(Achieved using matlab digital watermarking, which uses the technology is LSB)

minimizer of in-line coded functions

说明：  一种基于短时傅里叶时频分析的盲源分离算法，算法基于卷积混合。处理非平稳源信号。(Short Time Fourier based time-frequency analysis of blind source separation algorithm, hybrid algorithm is based on convolution. Handling non-stationary source signals.)

说明：  通过利用Weichselberger模型产生瑞利衰落信道(rayleigh)

该融合工具箱由一系列保存成.m的格式文件函数组成，可对经过配准好的灰度图像进行融合操作。界面操作简单，在界面上可以对相关参数进行设置。(The integration kit is saved into a format function of m, operating after registration good grayscale image fusion. The interface is simple, set the relevant parameters in the interface.)

This project seeks to implement H.264 (the standard x264 is based on) to run on CUDA. Specifically we attempt to write the motion estimation module in CUDA. Motion estimation is currently the most computationally intensive portion of the H.264 encoding process. Previous attempts to parallelize this algorithm in CUDA showed sizeable speedups but sacrificed quality by ignoring the motion vector prediction (MVp). We demonstrate the viability of a hierarchical (pyramid) motion estimation algorithm in CUDA. This solution addresses the MVp problem while still taking advantage of the CUDA framework.