搜索资源列表
BAM_NN
- 用外积和法设计的权矩阵,不能保证p对模式全部正确的联想。若对记忆模式对加以限制(即要求p个记忆模式Xk是两两正交的),则用外积和法设计的BAM网具有较好的联想能力。 在难以保证要识别的样本(或记忆模式)是正交的情况下,如何求权矩阵,并保证具有较好的联想能力?这个问题在用BAM网络实现对字符的识别程序仿真中得到体现。我们做过尝试,用伪逆法求权矩阵,虽然能对未加干扰的字符全部进行识别,但对加有噪声的字符识别效果很差。至于采用改变结构和其他算法的方法来求权矩阵,将是下一步要做的工作。-foreign
QAMTransmissionThroughaCompanding
- Abstract-The effect of the companding process on QAM signals has been under investigation for the past several years. The compander, included in the PCM telephone network to improve voice performance, has an unusual affect on digital QAM data
Advanced.Signal.Processing.and.Noise.Reduction.2nd
- Signal processing theory plays an increasingly central role in the development of modern telecommunication and information processing systems, and has a wide range of applications in multimedia technology, audio-visual signal processing, cellular mob
ghmm470
- 对具有随机噪声的二阶系统的模型辨识,进行标幺化以后系统的参考模型差分方程为: y(k)=a1*y(k-1)+a2*y(k-2)+b*u(k-1)+s(k) 式中,a1=0.3366,a2=0.6634,b=0.68,s(k)为随机噪声。由于神经网络的输出最大为1,所以,被辨识的系统应先标幺化,这里标幺化系数为5。采用正向建模(并联辨识)结构,神经网络选用3-9-9-1型,即输入层i,隐层j包括2级,输出层k的节点个数分别为3、9、9、1个;由于神经网络的最大输出为1,因此在辨识前应对原系统参考模
szsb_src
- VC++基于神经网络的数字图片识别技术,并可对图片进行灰度处理、二值化、递推锐化、去离噪声、字符分隔等处理功能,调试时请将在Debug目录中生成的EXE文件拷贝至Release目录里运行,因为那里有测试图片。 -VC++ based on neural network digital image recognition technology, and gray on the picture processing, binarization, recursive sharpening to a
szsb
- VC++基于神经网络的数字图片识别技术,并可对图片进行灰度处理、二值化、递推锐化、去离噪声、字符分隔等处理功能,调试时请将在Debug目录中生成的EXE文件拷贝至Release目录里运行,因为那里有测试图片。-VC++ based on neural network digital image recognition technology, and gray on the picture processing, binarization, recursive sharpening to away
example
- circles Various circle equations gain Constant gain circles noise Constant noise circles quality Constant Qn circles stability Stability circles conversion Conversion routines between different twoport network representations global Som
Lab1-Windows1
- 利用所学数据链路层原理,自己设计一个滑动窗口协议,在仿真环境下编程实现有噪音信道环境下两站点之间无差错双工通信。信道模型为8000bps全双工卫星信道,信道传播时延270毫秒,信道误码率为10-5,信道提供字节流传输服务,网络层分组长度固定为256字节。 通过该实验,进一步巩固和深刻理解数据链路层误码检测的CRC校验技术,以及滑动窗口的工作机理。滑动窗口机制的两个主要目标:(1) 实现有噪音信道环境下的无差错传输 (2)充分利用传输信道的带宽。在程序能够稳定运行并成功实现第一个目标之后,运
RKX-750-6B
- RKX-750-6B是一款经济型的超低单向接收机,适用于光纤进楼及农村中偏远的住房使用。采用国际先进的GaAs HBT集成电路技术,使光接收机具有非常低的噪音功率。在相同接收光功率的条件下,CNR要比普通的光接收机提高至少5.0dB以上,在接收光功率达-6dB时,系统的载噪比CNR仍然维持在50dB以上,从而节省发送光功率。特别是在接收机光功率较低的情况下,效果更为显著。 使用超低光功率接收机的技术,可以用比较小的投资延伸光纤传输长度,也为构造大型的光纤网络寻找了一种新的方法。一方面节省了发
ceshi
- 以实际所测随机中子脉冲数据的自相关函数为研究对象,借助仿真实验,开展Elman神经网络对不同浓度核材料进行识别的研究。在实测数据的基础上,通过叠加随机噪声,模拟产生了不同浓度核材料的相关函数样本用于神经网络的训练与测试,实验结果表明,-Calculating the stochastic neutron pulse autocorrelation function with the actually measured data as the research object, we carry o
TL_recongnize_add_noise8_8
- 加噪声的神经网络TL 识别。 将3*3网格中只有0,1两种状态,在识别时将矩阵转化为1*9的数组进行识别,加入噪声评估神经网络的学习能力。-TL neural network recognition plus noise. 3* 3 grid will only 0,1 two states, in identifying the matrix into an array of 1* 9 to identify, evaluate noise added learning ability o
final_Loadbalancing
- The demand for robust face recognition in real-world surveillance cameras i ncreasing due to the needs of practical applications such as security and surveillance. Although face recognition has been studied extensively in the literature, achiev
funnao
- 包括调制,解调,信噪比计算,光纤无线通信系统中传输性能的研究,采用加权网络中节点强度和权重都是幂率分布的模型。- Includes the modulation, demodulation, signal to noise ratio calculation, Fiber Transmission wireless communication system performance, Using weighted model nodes in the network strength and we
laonan_v54
- 添加噪声处理,采用加权网络中节点强度和权重都是幂率分布的模型,复化三点Gauss-lengend公式求pi。- Add noise processing, Using weighted model nodes in the network strength and weight are power law distribution, Complex of three-point Gauss-lengend the Formula pi.
baisan_v20
- 采用的是脉冲对消法,采用加权网络中节点强度和权重都是幂率分布的模型,利用matlab写成的窄带噪声发生。- It uses a pulse of consumer law, Using weighted model nodes in the network strength and weight are power law distribution, Using matlab written narrowband noise occurs.
bp2
- 采用贝叶斯正则化算法提高 BP 网络的推广能力。 在本例中,我们采用两种训练方法,即 L-M 优化算法(trainlm)和贝叶斯正则化算法(trainbr),用以训练 BP 网络,使其能够拟合某一附加有白噪声的正弦样本数据。-Bayesian regularization algorithm to improve the generalization ability of BP network. In this example, we use two training methods, na
sanfan
- 采用加权网络中节点强度和权重都是幂率分布的模型,一种噪声辅助数据分析方法,有CDF三角函数曲线/三维曲线图。- Using weighted model nodes in the network strength and weight are power law distribution, A noise auxiliary data analysis method, There CDF trigonometric curve/3D graphs.
mccullotch-pitt-matlab-code
- The following Matlab project contains the source code and Matlab examples used for neural network (mlp) robot localization. Comparison with ground truth and triangulation provided, with varying amounts of gaussian noise added in train and test data.
adversarial.tar
- 此程序为对抗生成网络,生成图像。 生成对抗网络是一种生成模型(Generative Model),其背后基本思想是从训练库里获取很多训练样本,从而学习这些训练案例生成的概率分布。 而实现的方法,是让两个网络相互竞争,‘玩一个游戏’。其中一个叫做生成器网络( Generator Network),它不断捕捉训练库里真实图片的概率分布,将输入的随机噪声(Random Noise)转变成新的样本(也就是假数据)。另一个叫做判别器网络(Discriminator Network),它可以同时观察真实和假