bdfm_1

针对信号自适应chirplet分解未知参数多、实现起来比较困难的特点，文献[1]提出了一种新的chirplet分解快速算法。该算法利用计算信号的二次相位函数，得到其能量分布集中于于信号的调频率曲线上的结论，此时通过谱峰检测可同时获得chirplet调频率、时间中心和幅度的估计；然后通过解线性调频技术获得其初始频率和宽度的估计，仿真结果验证了本文算法的有效性。 (For the signal adaptive the chirplet decomposition of unknown parameters to achieve difficult to literature [1] proposed a new fast algorithm for chirplet decomposition. The algorithm makes use of the secondary phase of the calculated signal function, the conclusion focused on energy distribution curve to tune the frequency of the signal by the peak detection chirplet tune frequency, time center and amplitude estimates then by solving chirp technology to obtain the initial frequency and width is estimated that the simulation results demonstrate the effectiveness of the proposed algorithm.)

matlab code for Generate maximally perceptually-distinct colors

说明：  基于遗传算法的图像阈值分割。里面有代码和本人搜集的参考文献，有助于初学者学习。(Based on genetic algorithm for image thresholding. I gathered codes and references which are very helpful for beginners.)

Using the PSO the LSSVM the parameter, the optimal parameters, more accurate classification.

BP神经网络容易陷于局部极小值，PSO算法在无约束非线性函数优化方面性能优越，通常可以直接找寻到全局最优解，即使不能搜多到全局最优解，也距离全局最优点不远。当然，基本PSO算法陷入局部极值也是有的。对于这个缺点目前还没有找到比较有效、省市的解决方案。本案例实现利用PSO算法和BP算法共同训练神经网络，先将网络进行PSO算法训练，然后BP算法接着进行小范围精细搜索，PSO算法训练神经网络的本质就是将输出误差函数（即能量函数）看成目标函数，PSO对能量函数进行全局寻找最小值。(BP neural networks are prone to local minimum values. The PS algorithm has superior performance in the optimization of unconstrained nonlinear functions. It can usually find the global optimal solution directly. Even if it can not find more global optimal solutions, it is not far from the global best. Of course, there is also a local extreme of the basic PO algorithm. For this shortcoming, there is no more effective, provincial and municipal solution. This case realizes the use of the SO algorithm and the BP algorithm to train the neural network. First, the network is trained in the SO algorithm, and then the BP algorithm is followed by a small range of fine searches. The essence of the PO algorithm training neural network is to regard the output error function(ie, the energy function) as the objective function, and the PO seeks a global minimum value for the energy function.)

说明：  弗兰衍射的matlab模拟程序。 程序来源于网络 谢谢大家(Ferran diffraction Matlab simulation program. Proceedings from the network Thank you)

this file implement weighted linear method in pattern recognition

模型预测控制的一个小例子~Gravity Drained Water Tank(A little example in MPC ——Gravity Drained Water Tank)

说明：  此程序给出信号的多分辨率分析（Multiresolution Analysis），即信号的多分辨率分解（Multiresolution Decomposition）方法及应用。该程序是一个多抽样率系统（Multirate System）举例，用M个的滤波器，将信号的频谱等分成M部分，使所得抽样频率降为1/M，然后再依据它们的“重要性”给一不同的bit数，那么所传输的bit流必然会进一步下降，起到压缩带宽的作用。(This program gives the signal of the multi-resolution analysis (Multiresolution Analysis), the multi-resolution signal decomposition (Multiresolution Decomposition) method and its application. The program is a multirate rate system (Multirate System) example, with M a filter, the signal' s frequency spectrum is divided into M parts, the proceeds of sampling frequency reduced to 1/M, then according to their &quot importance&quot to the bit of a different number, then the transmitted bit stream is bound to decline further, to play the role of bandwidth compression.)

matlab code for Download Daily Data Google and Yahoo! Finance