eigtoollib

说明：  伪随机码或称PN码，是扩频通信中不可缺少的一种序列。扩频通信，简单的说来就是通过一组相关性很强的伪随机序列，使原有的信号按照伪随机序列的特性出现。从波形上看，原有信号完全淹没于随机的噪声中，达到隐藏信号和抗干扰的目的。接受方只要有相同的伪随机序列，就可以通过伪随机序列很强的相关性将信号还原出来。而m序列，是扩频系统中最长用的伪随机序列。下面的代码就是基于matlab的生成m序列的代码，通过改变初始多项式的值，就可以生成不同阶数的m序列。(Pseudo-random code or PN code, spread spectrum communication is indispensable to a sequence. Spread spectrum communication, a simple to say that the adoption of a group of highly related to pseudo-random sequence, so that the original signal in accordance with the characteristics of pseudo-random sequence there. From the wave point of view, the original signal completely submerged in the random noise to hide the purpose of signal and interference. As long as the recipient has the same pseudo-random sequence, the pseudo-random sequence can be a strong correlation to restore the signal out. And m sequence spread spectrum system is the longest of the pseudo-random sequence used. The following code is generated based on m sequences of matlab code, by changing the value of the initial polynomial, we can generate a different order of the m sequence.)

CInterface simulation in matlab simulink

用Simulink中NCO模块产生正弦信号的程序，自己写的，采样率180MHz，正弦频率70MHz，可以直接运行。(Using Simulink module generates a sinusoidal signal of NCO program, sample rate 180MHz, sinusoidal frequency 70MHz, can be directly run.)

说明：  深度神经网络在函数近似中表现优越，然而需要从头开始训练。另一方面，贝叶斯方法，像高斯过程（GPs），可以利用利用先验知识在测试阶段进行快速推理。然而，高斯过程的计算量很大，也很难设计出合适的先验。本篇论文中我们提出了一种神经模型，条件神经过程（CNPs），可以结合这两者的优点。CNPs受灵活的随机过程的启发，比如GPs，但是结构是神经网络，并且通过梯度下降训练。CNPs通过很少的数据训练后就可以进行准确的预测，然后扩展到复杂函数和大数据集。我们证明了这个方法在一些典型的机器学习任务上面的的表现和功能，比如回归，分类和图像补全(Deep neural networks perform well in function approximation, but they need to be trained from scratch. On the other hand, Bayesian methods, such as Gauss Process (GPs), can make use of prior knowledge to conduct rapid reasoning in the testing stage. However, the calculation of Gauss process is very heavy, and it is difficult to design a suitable priori. In this paper, we propose a neural model, conditional neural processes (CNPs), which can combine the advantages of both. CNPs are inspired by flexible stochastic processes, such as GPs, but are structured as neural networks and trained by gradient descent. CNPs can predict accurately with very little data training, and then extend to complex functions and large data sets. We demonstrate the performance and functions of this method on some typical machine learning tasks, such as regression, classification and image completion.)

基于MH算法的贝叶斯模型更新，结构为五层框架结构，更新得到结构的物理参数(The Bayesian model based on the MH algorithm is updated and the structure is a five-story frame structure. The physical parameters of the structure are updated)

采用模糊C均值对数据集data聚为cluster_n类(Fuzzy C-means data sets of data together for cluster_n category)

考虑了基站通过多个中继与多个用户的问题，基于背包问题提出了最优功率分配。(the optimal power allocation is presented based on knapsack problem for a system with a base station communication with multiple users assisted by multiple relay.)

RLS算法，用matlab实现，用自适应算法来实现均衡，程序正确(RLS algorithm, using matlab to achieve, with an adaptive algorithm to achieve a balanced and correct procedures)

针对图片或视频数据的眼睛区域的检测，以及眨眼的检测确定人脸区域的程序。 份关于智能图像增强的美国专利，对图像处理感兴趣的朋友可以看一下。(Tracking learning)

GMRES算法，求解大型稀疏线性方程的快速数值解法，方便实时求解(GMRES algorithm is prosed to solve Large- scale sparse linear equations)