Nonlinear time series modelling and prediction using Gaussian RBFnetworks with enhanced clustering and RLS learning

An improved clustering and recursive least squares (RLS) learning algorithm for Gaussian radial basis function (RBF) networks is described for modelling and predicting nonlinear time series. Significant performance gain can be achieved with a much smaller network compared with the usual clustering and RLS method     

多尺度高斯模型的短期负荷预测

为了准确地进行电力负荷的短期预测,借鉴小波分析中对函数进行多尺度表示的思想,文中在高斯过程模型的基础上提出了多尺度高斯过程模型.通过选择合适的尺度参数,采用计算预测均方误差值大小的策略获取最佳延迟时间和最优嵌入维数对,然后对西北某地区电力系统进行短期负荷预测.与传统的支持向量机、径向基函数网负荷预测方法相比,基于多尺度高斯过程模型的短期负荷预测方法预测精度与支持向量机方法相当,性能优于径向基函数.     

基于HBF神经网络的自适应观测器

传统的RBF(Radial Basis Function)神经元基函数通常把高斯类型与单一宽度作为每个神经元的激活函数,这些特性限制了网络神经元的性能,特别是在处理复杂的非线性建模问题上.为了克服这个限制,本文应用了具有类似RBF网络,但激活函数不同-超基函数HBF(Hyper Basis Function)的网络.结合RBF网络,分析了HBF网络的结构、基函数形式及基函数对网络的影响,利用决策树算法计算了网络中心.在此基础上,提出了一种基于HBF神经网络的自适应观测器设计方法,并通过引入Lyapunov函数,证明了这种观测器设计方法的稳定性;最后通过仿真验证了这种HBF神经网络观测器能很好地观测系统的状态值.     

High-quality voice conversion system based on GMM statistical parameters and RBF neural network

A voice conversion （VC） system was designed based on Gaussian mixture model （GMM） and radial basis function （RBF） neural network. As a voice conversion model, RBF network needs quantities of training data to improve its performance. For one speech, the networks trained by different segments of data have different transformation effects. Since trying segment by segment to obtain the best conversion effect is complex, a conversion method was proposed, that uses GMM for statistics before training RBF network to aim at the problem. The speech transformation and representation using adaptive interpolation of weighted spectrum （STRAIGHT） model is used for accurate extraction of vocal tract spectrum. Then GMM is used to classify the numerous spectral parameters. The obtained mean parameters were trained in RBF network. Experiment reveals that, the soft classification ability of GMM can promptly realize the reduction and classification of training data under the premise of ensuring the training effect. The selection complexity is decreased thereafter. Compared to the conventional RBF network training methods, this method can make the transformation of spectral parameters more effective and improve the quality of converted speech.     

Comparing support vector machines with Gaussian kernels to radialbasis function classifiers

The support vector (SV) machine is a novel type of learning machine, based on statistical learning theory, which contains polynomial classifiers, neural networks, and radial basis function (RBF) networks as special cases. In the RBF case, the SV algorithm automatically determines centers, weights, and threshold that minimize an upper bound on the expected test error. The present study is devoted to an experimental comparison of these machines with a classical approach, where the centers are determined by X-means clustering, and the weights are computed using error backpropagation. We consider three machines, namely, a classical RBF machine, an SV machine with Gaussian kernel, and a hybrid system with the centers determined by the SV method and the weights trained by error backpropagation. Our results show that on the United States postal service database of handwritten digits, the SV machine achieves the highest recognition accuracy, followed by the hybrid system. The SV approach is thus not only theoretically well-founded but also superior in a practical application     

Supervised and unsupervised learning in radial basis functionclassifiers

The paper considers a number of strategies for training radial basis function (RBF) classifiers. A benchmark problem is constructed using ten-dimensional input patterns which have to be classified into one of three classes. The RBF networks are trained using a two-phase approach (unsupervised clustering for the first layer followed by supervised learning for the second layer), error backpropagation (supervised learning for both layers) and a hybrid approach. It is shown that RBF classifiers trained with error backpropagation give results almost identical to those obtained with a multilayer perceptron. Although networks trained with the two-phase approach give slightly worse classification results, it is argued that the hidden-layer representation of such networks is much more powerful, especially if it is encoded in the form of a Gaussian mixture model. During training, the number of subclusters present within the training database can be estimated: during testing, the activities in the hidden layer of the classification network can be used to assess the novelty of input patterns and thereby help to validate network outputs     

基于CMOS模拟电路的径向基函数神经网络

提出了可构成径向基函数(RBF)神经网络的CMOS模拟单元电路,包括绝对值电路、电流求均方根电路和类高斯函数电路.基于这些电路设计了一个二输入/一输出,含有两个隐层神经元的径向基函数神经网络,并通过异或问题进行了验证.所有单元均采用HJTC 0.18μm CMOS数模混合工艺制造,芯片面积为200μm×150μm,功耗约为100μW.芯片测试结果表明:提出的单元电路结构简单,功耗低,便于扩展和调节,因而为硬件实现径向基函数神经网络的片上学习提供了可     

基于自适应RBFN阵列的说话人识别

提出一种可用于说话人识别的自适应RBFN阵列。RBF网设计的核心在于确定网络中心的数目及位置，该自适应算法有效地融合了IOC与ROLS算法的优点，不仅能动态调节RBF网的隐节点数，还能使网络的数据中心自适应变化，很好地优化了网络的结构。用与文本无关的闭集说话人识别系统对该算法进行了验证，实验结果表明，该方法与传统的RBF算法相比，自适应RBF网具有较好的鲁棒性以及精简的网络结构等优点。     

基于PSO-RBF神经网络的模拟电路故障诊断研究

针对模拟电路故障诊断的难点和传统诊断方法的不足之处,提出了一种基于PSO算法优化的RBF神经网络模拟电路故障诊断方法。为了约简网络结构从而提高诊断效率,采用主成分分析方法对故障特征进行有效提取。针对RBF网络传统训练算法中隐层节点中心及基函数宽度选取困难问题,提出采用PSO算法来优化训练RBF网络,以提高网络的训练速度和泛化性能。最后,通过电路仿真对所提方法的有效性进行了验证。     

Neural Computation of the MoM Matrix Elements for Planar Configurations

In this paper a new artificial neural network (ANN) based model for the calculation of the method of moments (MoM) matrix elements is presented. Training sets that characterize the matrix elements are first constructed. These sets are then utilized to effectively train two radial basis function (RBF) neural networks to accurately estimate all the elements of the MoM matrix for any mesh used. The potential of the proposed approach is demonstrated in the case of a narrow microstrip line. The current distribution on the microstrip line produced by the trained RBF networks agrees very well with the exact distribution. In addition, the proposed ANN model is much faster than the conventional MoM procedure.     

Extended adaptive RBF equaliser for overcoming co-channelinterference

The authors investigate the problem of nonlinear adaptive equalisation in the presence of intersymbol interference, additive white Gaussian noise and co-channel interference. An extended radial basis function (RBF) network is proposed, in which regression weights are used in the output layer and the hidden unit is defined to have the Gaussian formula with the Mahalanobis distance. It is shown by simulation that the proposed structure gives reduced computational complexity without performance degradation, compared to that of the conventional RBF equaliser     

Neural network techniques for adaptive multiuser demodulation

Adaptive methods for performing multiuser demodulation in a direct-sequence spread-spectrum multiple-access (DS/SSMA) communication environment are investigated. In this scenario, the noise is characterized as being the sum of the interfering users' signals and additive Gaussian noise. The optimal receiver for DS/SSMA systems has a complexity that is exponential in the number of users. This prohibitive complexity has spawned the area of research on suboptimal receivers with moderate complexity. Adaptive algorithms for detection allow for reception when the communication environment is either unknown or changing. Motivated by previous work with radial basis functions (RBF's) for performing equalization, RBF networks that operate with knowledge of only a subset of the system parameters are studied. Although this form of detection has been previously studied (group detection) when the system parameters are known, in this work, neural network techniques are employed to adaptively determine unknown system parameters. This approach is further bolstered by the fact that the optimal detector in the synchronous case can be implemented by a RBF network when all of the system parameters are known. The RBF network's performance (with estimated parameters) is compared with the optimal synchronous detector, the decorrelating detector and the single layer perceptron detector. Clustering techniques and adaptive least mean squares methods are investigated to determine the unknown system parameters. This work shows that the adaptive radial basis function network attains near optimal performance and is robust in realistic communication environments     

径向基函数神经网络的软竞争学习算法

本文构造了径向基函数(RBF)神经网络的一类软竞争学习算法(SCLA).该算法的主要思想是首先在高斯基函数中心向量的训练过程中引入了隶属度函数,对每个输入样本,所有中心向量根据该样本属于其代表的类的隶属度值的大小进行自适应地调整;第二,把隶属度函数的模糊因子的倒数与模拟退火算法中的温度等同起来,在迭代过程中采用递增的方式来调整它.SCLA是RBF网络基于k-均值方法训练中心向量的学习算法的软竞争格式,它可以克服后者对初始值敏感和死节点的问题.仿真实验论证了SCLA是有效的.     

基于极速学习的粗糙RBF神经网络

提出了一种用于训练粗糙RBF神经网络(rough RBF neural networks,R-RBF)的极速学习机(extreme learning machine,ELM)方法,通过引入矩阵的Moore-Penrose逆,将传统的迭代学习方法转换为一种求线性方程的极小范数最小二乘解的方法.实验证明,在训练精度、训练时间上都能够达到非常优越的性能,其泛化精度能够提升50%以上.     

基于递归正交最小二乘的径向基函数网络人脸识别

本文提出了一种基于递归正交最小二乘的径向基函数(RBF)网络人脸识别算法,该算法首先使用主成分分析(PCA)提取输入图像特征,将提取的特征作为RBF网络的输入进行识别,在求取网络权值时采用递归正交最小二乘(ROLS)算法。实验表明,该算法能明显地缩短训练时间同时具有较高的识别率。     

组合二进制偏移载波信号的伪码周期及组合码序列盲估计

针对在非协作通信以及低信噪比下组合二进制偏移载波(CBOC)信号伪码周期和组合码序列较难估计的问题,该文提出了2次谱算法与基于径向基函数(RBF)神经网络算法。对输入信号进行2次功率谱计算,可以得到CBOC信号的伪码周期。在此基础上,首先对接收的1周期组合码序列进行重叠分段,其次优化筛选出学习系数,对每段数据向量作为RBF神经网络的输入信号并进行有监督地调节,最后对每段数据向量多次输入并反复训练权值向量就可以恢复原组合码序列。仿真结果表明,利用2次谱可以在低信噪比下估计出伪码周期;在误码率低于1%的情况下,所提出的RBF神经网络相比于反向传播(BP)神经网络与Sanger神经网络,信噪比分别提高1 dB和3 dB,并且在同等条件下所需的数据组数较少。

     

基于小波变换和多神经网络融合的雷达一维距离像目标识别

针对低信噪比情况下目标的雷达一维距离像性能较差的问题,提出利用小波去噪以增强目标一维距离像的方法.对去噪后的目标一维距离像提取归一化中心矩,分别采用径向基函数网络和BP网络进行分类识别;在分析了两类网络用于分类识别特点基础上,指出具有不同拓扑结构和传递函数的神经网络对于各类训练样本分布状况的学习和描述能力不同,提出了一种综合利用两种网络做融合识别的新方法.通过对5类飞机暗室测量数据的实验,验证了上述方法的有效性.     

无线电信号识别方法的研究

文章提出了小波分析与神经网络相结合的方法自动识别无线电信号的调制类型.文中分别用常规的方法和小波的方法提取信号的特征参数，送入RBF（Radial Basis Function）网络，按照样本距离最小的原则进行聚类，利用RBF网络的快速收敛性和较好的自适应性，实现对无线电信号的识别.仿真结果表明，采用小波与神经网络相结合的分类方法，能获得满意的识别率。     

RBF神经网络在视频业务建模及预测中的应用

ＡＴＭ技术在关键在于业务控制,能否有效地实施业务控制则取决于对业务特征的了解和预测能力。传统的解析方法对视频业务进行预测的局限性较为明显。本文采用径向基函数神经网络对视频业务进行建模和预测,并提出了分别采用ＬＢＧ算法和Ｈｅｓｔｅｎｅｓ奇异值解进行隐含层神经元中心选择和输出神经腱权值计算的改进方法。