完美的无迹卡尔曼滤波在目标跟踪开源程序

介绍了Gabor的几种形式的函数以及其实验图像(The introduction of several forms of the Gabor function, as well as its experimental images)

This simulink model simulates the damped driven pendulum, showing it s chaotic motion. theta = angle of pendulum omega = (d/dt)theta = angular velocity Gamma(t) = gcos(phi) = Force omega_d = (d/dt) phi Gamma(t) = (d/dt)omega + omega/Q + sin(theta) Play with the initial conditions (omega_0, theta_0, phi_0 = omega(t=0), theta(t=0), phi(t=0)) and the system parameters (g, Q, omega_d) and the solver parameters/method. Chaos can be seen for Q=2, omega_d=w/3. The program outputs to Matlab time, theta(time) & omega(time). Plot the phase space via: plot(mod(theta+pi, 2*pi)-pi, omega, . ) Plot the Poincare sections using: t_P = (0:2*pi/omega_d:max(time)) plot(mod(spline(time, theta+pi, t_P), 2*pi)-pi, spline(time, omega, t_P), . ) System is described in: "Fractal basin boundaries and intermittency in the driven damped pendulum" E. G. Gwinn and R. M. Westervelt PRA 33(6):4143 (1986) (This simulink model simulates the damped driven pendulum, showing it s chaotic motion. theta = angle of pendulum omega = (d/dt)theta = angular velocity Gamma(t) = gcos(phi) = Force omega_d = (d/dt) phi Gamma(t) = (d/dt)omega+ omega/Q+ sin(theta) Play with the initial conditions (omega_0, theta_0, phi_0 = omega(t=0), theta(t=0), phi(t=0)) and the system parameters (g, Q, omega_d) and the solver parameters/method. Chaos can be seen for Q=2, omega_d=w/3. The program outputs to Matlab time, theta(time) & omega(time). Plot the phase space via: plot(mod(theta+pi, 2*pi)-pi, omega, . ) Plot the Poincare sections using: t_P = (0:2*pi/omega_d:max(time)) plot(mod(spline(time, theta+pi, t_P), 2*pi)-pi, spline(time, omega, t_P), . ) System is described in: "Fractal basin boundaries and intermittency in the driven damped pendulum" E. G. Gwinn and R. M. Westervelt PRA 33(6):4143 (1986) )

A new blind adaptive multiuser detection scheme based on a hybrid of Kalman filter and subspace estimation is proposed. It is shown that the detector can be expressed as an anchored signal in the signal subspace and the coefficients can be estimated by the Kalman filter using only the signature waveform and the timing of the desired user.(A new blind adaptive multiuser detection s cheme based on a hybrid of the Kalman filter and subs pace estimation is proposed. It is shown that th e detector can be expressed as an anchored signa l in the signal subspace and the coefficients ca n be estimated by the Kalman filter using only th e signature waveform and the timing of the desir ed user.)

说明：  信号SG平滑滤波Matlab程序仿真，简单实用，对信号的前期处理很有用(SG signal smoothing filter Matlab simulation program, simple and practical, pre-treatment of the useful signal)

该算法是在 MATLAB GUI 中实现并插入到通过函数字段创建仿真环境。

pso粒子群优化算法，适用于计算目标函数，全局非线性反演的问题(pso Particle Swarm Optimization for calculating the objective function, the global nonlinear inversion problem)

Programmi che descrivono l utilizzo delle CNN

提取Gabor特征，LBP特征，LGBP特征代码(extract Gabor feature,LBP feature and LGBP feature.)

说明：  摘要：用MATLAB对基带数据传输解调检测进行仿真，推导不同先验概率下，接收端最小错误的判决门限。仿真不同先验概率下的误码性能，二进制与多进制的误码性能比较，码间串扰对误码性能的影响，及频偏对串行通信误码率的影响。 关键词：MATLAB，基带传输，检测，判决门限 (Abstract: Using MATLAB for baseband demodulation simulation data derived under different prior probability, the receiver decision threshold minimum error. Simulation under different prior probability of error performance, binary and multi-bit error performance of binary comparison ISI on the BER performance, and the offset of the serial communication error rate. Keywords: MATLAB, baseband transmission, detection, decision threshold)

用镜像矩阵求出矩阵的QR分解（正交三角分解），并求解线性方程组，该方法不必选主元，但其计算过程非常稳定。也可用于求矩阵的广义逆和求解线性最小二乘问题。子过程QRDCMP用镜像矩阵求m*n矩阵A的QR分解，A=Q^TR,其中Q是m*m正交矩阵，R为m*n上三角矩阵；子过程QRBKSB用矩阵的QR分解求解线性方程组Ax=b，其中A为n阶非奇异方阵。(Matrix obtained with the mirror matrix QR decomposition (orthogonal triangular decomposition), and solving linear equations, the method does not have to choose the main element, but its calculation process is very stable. Can also be used to find the generalized inverse matrix and solving linear least squares problems. Sub-process of the mirror matrix QRDCMP m* n matrix A QR decomposition, A = Q ^ TR, wherein Q is a m* m orthogonal matrix, R is m* n upper triangular matrix subprocess QRBKSB a QR decomposition to solve the matrix linear equations Ax = b, where A is non-singular square matrix of order n.)