mmse

最短路这个说明倒不是说明函数的功能，只是声明一下函数中出现的问题。 shortest()函数没有问题，在secshortest中我希望只得到一条次短路（或几条roadcost()值相同的次短路），结果却出来好几条，我仔细检查了一下程序，应该没有什么问题。 因此在使用时，要得到次短路，只能用roadcost()函数分别求一下各条路的权重和，找到最小的。 有些奇怪 (short path)

Architecture books Important ones.

This document is about zsource multilevel inverter.

Monte Carlo Simulation On Estimating the probability of a flush

提出了一个采用(2,1,7)卷积码+QPSK的中频调制解调方案,并在Xilinx公司的100万 门FPGA芯片上实现了该系统。该系统在信噪比SNR为6dB左右时可实现速率超过1Mbit/s、误码率 小于10-5的数据传输。 (Proposed a use of (2,1,7) convolutional code+ QPSK modulation and demodulation of the IF program, and in Xilinx' s FPGA chip one million on implementation of the system. The system SNR to 6dB signal to noise ratio at about the rate may achieve more than 1Mbit/s, less than 10-5 bit error rate of data transmission.)

说明：  这是一个c++实现的基本遗传算法 很好用的，只要按照提示输入数值区间就行了或者自己把那个函数改一下就可以做了。(c++ to achieve a good basic genetic algorithm used)

number of frames in a video for different scales

利用matlab里面的吹风机数据，用前300个数据进行辨识。分别采用最小二乘辨识和相关性系数法辨识。并画出图形。通过对比最小二乘法和相关函数两种方法得出来的辨识图形。可知相关函数辨识的准确度更高。(Use a hair dryer inside matlab data before using 300 data identification. Were used to identify the least squares identification and correlation coefficient method. And draw graphics. By comparing the least squares method and related functions have two ways out of the identification graphics. Shows a higher correlation function identification accuracy.)

关于高斯噪声去除的目前最优秀的算法 BM3D 欢迎大家交流学习(Welcome About the best algorithm BM3D Gaussian noise removal the exchange of learning)

Gribovskaya, E., Khansari Zadeh, S. M. and Billard, A. (2010) Learning Nonlinear Multi-Variate Motion Dynamics for Real-Time Position and Orientation Control of Robotic Manipulators. International Journal of Robotics Research. [Infoscience]