xianboqi

卡尔曼滤波的故障诊断，才用卡尔曼滤波方法的故障诊断技术的matlab 与simulink 仿真代码(karman filter with design)

说明：  采用全有全无法进行用户均衡（UE）条件下的流量分配(trips distribution when user equilibrium with all-or-nothing method)

神经网络训练程序、测试程序以及验证程序整体(Neural network training procedures, test procedures, and the overall validation process)

说明：  主要是通过GA和PSO的全局搜索能力，用于改进BP网络的权值阈值(It is mainly used to improve the weight threshold of BP network)

将自适应信号处理中的RLS和LMS算法应用到信道均衡中去，用对比图和眼图研究性能。(Adaptive signal processing RLS and LMS algorithm is applied to channel equalization go, with comparison chart and eye research performance.)

用ica进行混合图像分离，里面含有用matlab编写的源程序(Separation of mixed images using ica, which contains the source code written with matlab)

研究了重频参差相参脉冲串的频率估计算法。通过脉内相关积累，提高了新序列的信噪比。对新序列的相位差按参差重数抽样平均，减小了等效的相位噪声方差，利用重频参差比解相位模糊，扩大了频偏允许范围，降低了算法的信噪比门限。分析了本算法实现相参频率估计的条件，推导了相应信噪比门限的解析表达式，指出了信噪比门限与信号样本总数、参差重数、参差比之阎的关系。仿真结果表明：上述结论是正确的，在满足信噪比门限条件下，频率估计的精度接近相参脉冲串频率估计的克拉美一罗限(CRLB)。(Frequency Estimation for Staggered Coherent Pulse Train)

prediction of voice character with BP neural network

卡尔曼滤波器的操作包括两个阶段:预测与更新。在预测阶段,滤波器使用上一状态的估计,做出对当前状态的估计。在更新阶段,滤波器利用对当前状态的观测值优化。(Kalman filter operation consists of two stages: prediction and update. In the prediction stage, the filter is estimated using the previous state, to make an estimate of the current state. In the update phase, the filter use observations on the current state of the optimization.)

退火粒子群优化算法（SAPSO），matlab代码，仅供学习(Annealing particle swarm optimization (SAPSO), matlab code, only to learn)