资源列表
boltzman_VC
- Boltzmannn机网络是Hinton等人在1985年将模拟退火算法引入到神经网络中，提出的，简称BM网络。-Boltzmannn computer network is Hinton and others in 1985 will be simulated annealing primer access to the neural network, the network referred to BM.
CPN_VC
- CPN网络，即对向传播网络，是将Kohonen特征映射网络与Grossberg基本竞争网络相结合，发挥各自特长的一种新型特征映射网络。-CPN network, the right to communication networks, Kohonen feature is the mapping of network and basic Grossberg competing networks combined play to their strong points of a new featu
series2parallel
- adsl 模块中的另一部分，Serial/Parallel, Parallel/Serial及Constellation Mapping部分。-adsl Module of the other part, Serial / Parallel, Parallel / Serial and Constellation Mapping part.
BAM_VC
- BAM网络即双向联想记忆网络，是一种可以记忆模式对的两层非线性反馈神经网络。-BAM network is a two-way associative memory networks, it is a memory of the two-mode nonlinear feedback neural network.
Kmeans_VC
- 鼎鼎大名的Kmeans的VC实现。广泛用于神经网络、模式识别领域。-the Kmeans the famous VC. Widely used in neural networks, pattern recognition.
adsl_a2d_d2a
- This module provides an introduction for our ELEC 301 DMT group project. 这是其中a2d模块部分-This module provides an introduction for o ur ELEC 301 DMT group project. This module is part of a2d
HopfieldNN0-9Recognise
- This code in this directory implements the binary hopfield network. Source code may be found in HOPNET.CPP. A sample training file is H7x8N4.trn. Sample test pattern files are: H7x8D4.TST, H5x8D7.TST, H5x8D7.TST and H5x8D9.TST, Output of the pr
bp_train_VC
- BP的训练方法.删除init_wj_hid_output与init_wj_input_hid可以进行新的学习!不然只是按老方式学习!这是一个最简单的BP算法异或问题!没有阈值，学习速度不可改变!下次会改进!如果同或只需改变输入的方式就可以!-BP's training methods. Delete init_wj_hid_output with init_wj _input_hid can conduct new learning! Otherwise, only the old wa
ZipDotNet
- TCompressionStream和TDecompressionStream-TCompressionStream and TDecompressionStr eam
signal11
- 2ask,4ask,2fsk,4fsk,2psk,4psk,16qam,gmsk,msk,pi/4qpsk,oqpsk.11种数字信号的生成。-2ask, 4ask, 2fsk, 4fsk, 2psk, 4psk, 16qam, gmsk. msk, pi/4qpsk, oqpsk.11 kinds of digital signal generation.
featrue7
- 提取信号七个基于瞬时信息的特征：零中心归一化瞬时幅度功率谱密度的最大值,零中心归一化瞬时幅度绝对值的标准偏差，零中心非弱信号段瞬时相位非线性分量绝对值标准偏差，零中心非弱信号段瞬时相位非线性分量标准偏差，零中心归一化的非弱信号段瞬时频率绝对值的标准偏差，一个信号段的归一化瞬时频率功率谱密度的最大值，根据信号 QPSK 和16QAM在 XI 轴投影的不同表现，提出特征参数。-extract information based on the instan eous features : In the
DWNN_nn
- 小波神经网络的分类识别，学习率因子，网络动量项因子，网络多分辨级数，网络平移参数 都可自己选。本例输入节点5，输出5。可识别五种信号。-wavelet neural network classification, the learning rate factor, network momentum contribution factor, network multiresolution series. Translation network parameters can be of his ow