duda-Pattern-Classification-Matlab

MATLAB中读写、处理科学数据文件格式NETCDF的程序(code related to the reading, writing and other process about NETCDF)

上传的matlab程序是股票市场的市场冲击成本程序，欢迎大家下载查阅！(the file is the matlab program of Impact Cost! )

1D CNN和2D CNN网络的仿真实现，包含完整的子函数分解，采用matlab编写(demo_cnn_auto- demo for Autonomous CNNs with a symmetric A template solve_cnn2d- solve a 2D CNN system, Euler method solve_cnn2d_multi- solve a 2D multi-layered CNN system, Euler method Utility routines make_cnn_params- check and update the CNN params structure make_cnn_params_multi- check and update the multi-layer CNN params structure make_cnn_time- check and update the CNN time structure step_ahead1- advance a 1D CNN state one step forward, Euler method step_ahead2- advance a 2D CNN state one step forward, Euler method unit_sat- unity saturation )

基于BP神经网络算法的非线性系统拟合，在matlab中实现BP神经网络非线性拟合算法。(BP neural network algorithm fitting of nonlinear systems)

实现产生伪随机序列的部件 —— 线性反馈移位寄存器单元。 SFlog2为扩频因子的底数为2的对数值，cycle为PN序列的周期，其值为2^SFlog2。initial_state为移位寄存器的初始状态，generator_polynomial_coefficient为生成PN序列所需的本原多项式，对应于移位寄存器的连接向量。(The realization of pseudo-random sequence generated components- LFSR unit. SFlog2 factor for the spread-spectrum base 2 of value, cycle for the PN sequence cycle, its value is 2 ^ SFlog2. initial_state shift register for the initial state, generator_polynomial_coefficient to generate the necessary PN sequence primitive polynomial, corresponding to the shift register connection vector.)

一个matlab优化小程序，老师布置的作业，值得一看。(Matlab optimize a small program, teacher assignments, worth a visit.)

在运载体运动情况下,提出了一种适用于捷联惯导系统(SINS)初始对准的新算法,该算法以惯性空 间为过渡参考基准,即建立了初始时刻惯性坐标系(i0系)和初始时刻捷联惯组惯性坐标系(ib0系)两个惯性坐标 系。将捷联惯组坐标系(b系)与导航坐标系(n系)之间的初始对准姿态矩阵的实现分解为三步骤:(1)通过地理 位置和初始对准时间求解i0系至n系的变换矩阵 (2)通过惯导比力方程变形,引入测速仪速度辅助,求解从ib0 系至i0系的常值变换矩阵 (3)使用姿态更新算法实时计算b相对于ib0系的变换矩阵。最后,载车运动环境下的 初始对准试验结果表明航向角对准精度达到了0.1b(1R)。 (On the moving base, a new initial alignment algorithm for strapdown inertial navigation system (SINS) is proposed. In this algorithm, two special inertial frames, the initia-l time inertial frame (i0frame) and the initia-l time SINS body inertial frame (ib0frame), are defined and are selected as transition reference frames. The calculation of initial alignment direction cosine matrix (DCM) between SINS body frame (bframe) and nav- igation reference frame (nframe) is expanded into three parts: (1) the DCM betweeni0frame andnframe is easily obtained via local geographical coordinates and initial alignment time (2) by expressing SINS specific force equation inbframe and by the aid of external velocity log, the constant DCM fromib0frame toi0frame is achieved (3) attitude updating using gyro samples gets the DCM betweenbframe andib0frame. In the end, some vehicular tests are carried out on the moving base and the accuracy of azimuth reaches about 0.1b(1R). )

CSTR 系统仿真代码，用于故障诊断产生模拟数据(CSTR model)

本文就着重比较Radon-Wigner变换、离散Chirp-Fourier变换等方法的优劣。(This article on a comparison of the Radon-Wigner transformation, the advantages and disadvantages of the discrete Chirp-Fourier transform method.)

CFAR matlab simulation