Abstract: This paper uses the radial basis function neural
network (RBFNN) for system identification of nonlinear systems.
Five nonlinear systems are used to examine the activity of RBFNN in
system modeling of nonlinear systems; the five nonlinear systems are
dual tank system, single tank system, DC motor system, and two
academic models. The feed forward method is considered in this
work for modelling the non-linear dynamic models, where the KMeans
clustering algorithm used in this paper to select the centers of
radial basis function network, because it is reliable, offers fast
convergence and can handle large data sets. The least mean square
method is used to adjust the weights to the output layer, and
Euclidean distance method used to measure the width of the Gaussian
function.
Abstract: In this paper, a clustering algorithm named KHarmonic
means (KHM) was employed in the training of Radial
Basis Function Networks (RBFNs). KHM organized the data in
clusters and determined the centres of the basis function. The popular
clustering algorithms, namely K-means (KM) and Fuzzy c-means
(FCM), are highly dependent on the initial identification of elements
that represent the cluster well. In KHM, the problem can be avoided.
This leads to improvement in the classification performance when
compared to other clustering algorithms. A comparison of the
classification accuracy was performed between KM, FCM and KHM.
The classification performance is based on the benchmark data sets:
Iris Plant, Diabetes and Breast Cancer. RBFN training with the KHM
algorithm shows better accuracy in classification problem.
Abstract: This paper describes a method to improve the robustness of a face recognition system based on the combination of two compensating classifiers. The face images are preprocessed by the appearance-based statistical approaches such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). LDA features of the face image are taken as the input of the Radial Basis Function Network (RBFN). The proposed approach has been tested on the ORL database. The experimental results show that the LDA+RBFN algorithm has achieved a recognition rate of 93.5%
Abstract: A clustering based technique has been developed and implemented for Short Term Load Forecasting, in this article. Formulation has been done using Mean Absolute Percentage Error (MAPE) as an objective function. Data Matrix and cluster size are optimization variables. Model designed, uses two temperature variables. This is compared with six input Radial Basis Function Neural Network (RBFNN) and Fuzzy Inference Neural Network (FINN) for the data of the same system, for same time period. The fuzzy inference system has the network structure and the training procedure of a neural network which initially creates a rule base from existing historical load data. It is observed that the proposed clustering based model is giving better forecasting accuracy as compared to the other two methods. Test results also indicate that the RBFNN can forecast future loads with accuracy comparable to that of proposed method, where as the training time required in the case of FINN is much less.
Abstract: Facial expression analysis is rapidly becoming an
area of intense interest in computer science and human-computer
interaction design communities. The most expressive way humans
display emotions is through facial expressions. In this paper we
present a method to analyze facial expression from images by
applying Gabor wavelet transform (GWT) and Discrete Cosine
Transform (DCT) on face images. Radial Basis Function (RBF)
Network is used to classify the facial expressions. As a second stage,
the images are preprocessed to enhance the edge details and non
uniform down sampling is done to reduce the computational
complexity and processing time. Our method reliably works even
with faces, which carry heavy expressions.
Abstract: This paper presents a methodology towards the emulation of the electrical power consumption of the RF device during the cellular phone/handset transmission mode using the LTE technology. The emulation methodology takes the physical environmental variables and the logical interface between the baseband and the RF system as inputs to compute the emulated power dissipation of the RF device. The emulated power, in between the measured points corresponding to the discrete values of the logical interface parameters is computed as a polynomial interpolation using polynomial basis functions. The evaluation of polynomial and spline curve fitting models showed a respective divergence (test error) of 8% and 0.02% from the physically measured power consumption. The precisions of the instruments used for the physical measurements have been modeled as intervals. We have been able to model the power consumption of the RF device operating at 5MHz using homotopy between 2 continuous power consumptions of the RF device operating at the bandwidths 3MHz and 10MHz.
Abstract: The behavior of Radial Basis Function (RBF) Networks greatly depends on how the center points of the basis functions are selected. In this work we investigate the use of instance reduction techniques, originally developed to reduce the storage requirements of instance based learners, for this purpose. Five Instance-Based Reduction Techniques were used to determine the set of center points, and RBF networks were trained using these sets of centers. The performance of the RBF networks is studied in terms of classification accuracy and training time. The results obtained were compared with two Radial Basis Function Networks: RBF networks that use all instances of the training set as center points (RBF-ALL) and Probabilistic Neural Networks (PNN). The former achieves high classification accuracies and the latter requires smaller training time. Results showed that RBF networks trained using sets of centers located by noise-filtering techniques (ALLKNN and ENN) rather than pure reduction techniques produce the best results in terms of classification accuracy. The results show that these networks require smaller training time than that of RBF-ALL and higher classification accuracy than that of PNN. Thus, using ALLKNN and ENN to select center points gives better combination of classification accuracy and training time. Our experiments also show that using the reduced sets to train the networks is beneficial especially in the presence of noise in the original training sets.
Abstract: Earth reinforcing techniques have become useful and economical to solve problems related to difficult grounds and provide satisfactory foundation performance. In this context, this paper uses radial basis function neural network (RBFNN) for predicting the bearing pressure of strip footing on reinforced granular bed overlying weak soil. The inputs for the neural network models included plate width, thickness of granular bed and number of layers of reinforcements, settlement ratio, water content, dry density, cohesion and angle of friction. The results indicated that RBFNN model exhibited more than 84 % prediction accuracy, thereby demonstrating its application in a geotechnical problem.
Abstract: Background noise is particularly damaging to speech
intelligibility for people with hearing loss especially for sensorineural
loss patients. Several investigations on speech intelligibility have
demonstrated sensorineural loss patients need 5-15 dB higher SNR
than the normal hearing subjects. This paper describes Discrete
Cosine Transform Power Normalized Least Mean Square algorithm
to improve the SNR and to reduce the convergence rate of the LMS
for Sensory neural loss patients. Since it requires only real arithmetic,
it establishes the faster convergence rate as compare to time domain
LMS and also this transformation improves the eigenvalue
distribution of the input autocorrelation matrix of the LMS filter.
The DCT has good ortho-normal, separable, and energy compaction
property. Although the DCT does not separate frequencies, it is a
powerful signal decorrelator. It is a real valued function and thus
can be effectively used in real-time operation. The advantages of
DCT-LMS as compared to standard LMS algorithm are shown via
SNR and eigenvalue ratio computations. . Exploiting the symmetry
of the basis functions, the DCT transform matrix [AN] can be
factored into a series of ±1 butterflies and rotation angles. This
factorization results in one of the fastest DCT implementation. There
are different ways to obtain factorizations. This work uses the fast
factored DCT algorithm developed by Chen and company. The
computer simulations results show superior convergence
characteristics of the proposed algorithm by improving the SNR at
least 10 dB for input SNR less than and equal to 0 dB, faster
convergence speed and better time and frequency characteristics.
Abstract: The empirical mode decomposition (EMD) represents any time series into a finite set of basis functions. The bases are termed as intrinsic mode functions (IMFs) which are mutually orthogonal containing minimum amount of cross-information. The EMD successively extracts the IMFs with the highest local frequencies in a recursive way, which yields effectively a set low-pass filters based entirely on the properties exhibited by the data. In this paper, EMD is applied to explore the properties of the multi-year air temperature and to observe its effects on climate change under global warming. This method decomposes the original time-series into intrinsic time scale. It is capable of analyzing nonlinear, non-stationary climatic time series that cause problems to many linear statistical methods and their users. The analysis results show that the mode of EMD presents seasonal variability. The most of the IMFs have normal distribution and the energy density distribution of the IMFs satisfies Chi-square distribution. The IMFs are more effective in isolating physical processes of various time-scales and also statistically significant. The analysis results also show that the EMD method provides a good job to find many characteristics on inter annual climate. The results suggest that climate fluctuations of every single element such as temperature are the results of variations in the global atmospheric circulation.
Abstract: Monitoring lightning electromagnetic pulses (sferics) and other terrestrial as well as extraterrestrial transient radiation signals is of considerable interest for practical and theoretical purposes in astro- and geophysics as well as meteorology. Managing a continuous flow of data, automation of the analysis and classification process is important. Features based on a combination of wavelet and statistical methods proved efficient for this task and serve as input into a radial basis function network that is trained to discriminate transient shapes from pulse like to wave like. We concentrate on signals in the Very Low Frequency (VLF, 3 -30 kHz) range in this paper, but the developed methods are independent of this specific choice.
Abstract: In this paper, we consider the design of pulse shaping
filter using orthogonal Hermite-Rodriguez basis functions. The pulse
shaping filter design problem has been formulated and solved as a
quadratic programming problem with linear inequality constraints.
Compared with the existing approaches reported in the literature, the
use of Hermite-Rodriguez functions offers an effective alternative to
solve the constrained filter synthesis problem. This is demonstrated
through a numerical example which is concerned with the design of
an equalization filter for a digital transmission channel.
Abstract: The purpose of planned islanding is to construct a
power island during system disturbances which are commonly
formed for maintenance purpose. However, in most of the cases
island mode operation is not allowed. Therefore distributed
generators (DGs) must sense the unplanned disconnection from the
main grid. Passive technique is the most commonly used method for
this purpose. However, it needs improvement in order to identify the
islanding condition. In this paper an effective method for
identification of islanding condition based on phase space and neural
network techniques has been developed. The captured voltage
waveforms at the coupling points of DGs are processed to extract the
required features. For this purposed a method known as the phase
space techniques is used. Based on extracted features, two neural
network configuration namely radial basis function and probabilistic
neural networks are trained to recognize the waveform class.
According to the test result, the investigated technique can provide
satisfactory identification of the islanding condition in the
distribution system.
Abstract: The design of a complete expansion that allows for
compact representation of certain relevant classes of signals is a
central problem in signal processing applications. Achieving such a
representation means knowing the signal features for the purpose of
denoising, classification, interpolation and forecasting. Multilayer
Neural Networks are relatively a new class of techniques that are
mathematically proven to approximate any continuous function
arbitrarily well. Radial Basis Function Networks, which make use of
Gaussian activation function, are also shown to be a universal
approximator. In this age of ever-increasing digitization in the
storage, processing, analysis and communication of information,
there are numerous examples of applications where one needs to
construct a continuously defined function or numerical algorithm to
approximate, represent and reconstruct the given discrete data of a
signal. Many a times one wishes to manipulate the data in a way that
requires information not included explicitly in the data, which is
done through interpolation and/or extrapolation.
Tidal data are a very perfect example of time series and many
statistical techniques have been applied for tidal data analysis and
representation. ANN is recent addition to such techniques. In the
present paper we describe the time series representation capabilities
of a special type of ANN- Radial Basis Function networks and
present the results of tidal data representation using RBF. Tidal data
analysis & representation is one of the important requirements in
marine science for forecasting.
Abstract: The security of computer networks plays a strategic
role in modern computer systems. Intrusion Detection Systems (IDS)
act as the 'second line of defense' placed inside a protected
network, looking for known or potential threats in network traffic
and/or audit data recorded by hosts. We developed an Intrusion
Detection System using LAMSTAR neural network to learn patterns
of normal and intrusive activities, to classify observed system
activities and compared the performance of LAMSTAR IDS with
other classification techniques using 5 classes of KDDCup99 data.
LAMSAR IDS gives better performance at the cost of high
Computational complexity, Training time and Testing time, when
compared to other classification techniques (Binary Tree classifier,
RBF classifier, Gaussian Mixture classifier). we further reduced the
Computational Complexity of LAMSTAR IDS by reducing the
dimension of the data using principal component analysis which in
turn reduces the training and testing time with almost the same
performance.
Abstract: Purpose of this work is to develop an automatic classification system that could be useful for radiologists in the breast cancer investigation. The software has been designed in the framework of the MAGIC-5 collaboration. In an automatic classification system the suspicious regions with high probability to include a lesion are extracted from the image as regions of interest (ROIs). Each ROI is characterized by some features based generally on morphological lesion differences. A study in the space features representation is made and some classifiers are tested to distinguish the pathological regions from the healthy ones. The results provided in terms of sensitivity and specificity will be presented through the ROC (Receiver Operating Characteristic) curves. In particular the best performances are obtained with the Neural Networks in comparison with the K-Nearest Neighbours and the Support Vector Machine: The Radial Basis Function supply the best results with 0.89 ± 0.01 of area under ROC curve but similar results are obtained with the Probabilistic Neural Network and a Multi Layer Perceptron.
Abstract: Artificial Intelligence (AI) methods are increasingly being used for problem solving. This paper concerns using AI-type learning machines for power quality problem, which is a problem of general interest to power system to provide quality power to all appliances. Electrical power of good quality is essential for proper operation of electronic equipments such as computers and PLCs. Malfunction of such equipment may lead to loss of production or disruption of critical services resulting in huge financial and other losses. It is therefore necessary that critical loads be supplied with electricity of acceptable quality. Recognition of the presence of any disturbance and classifying any existing disturbance into a particular type is the first step in combating the problem. In this work two classes of AI methods for Power quality data mining are studied: Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs). We show that SVMs are superior to ANNs in two critical respects: SVMs train and run an order of magnitude faster; and SVMs give higher classification accuracy.
Abstract: This paper presents a novel methodology for Maximum Power Point Tracking (MPPT) of a grid-connected 20 kW Photovoltaic (PV) system using neuro-fuzzy network. The proposed method predicts the reference PV voltage guarantying optimal power transfer between the PV generator and the main utility grid. The neuro-fuzzy network is composed of a fuzzy rule-based classifier and three Radial Basis Function Neural Networks (RBFNN). Inputs of the network (irradiance and temperature) are classified before they are fed into the appropriated RBFNN for either training or estimation process while the output is the reference voltage. The main advantage of the proposed methodology, comparing to a conventional single neural network-based approach, is the distinct generalization ability regarding to the nonlinear and dynamic behavior of a PV generator. In fact, the neuro-fuzzy network is a neural network based multi-model machine learning that defines a set of local models emulating the complex and non-linear behavior of a PV generator under a wide range of operating conditions. Simulation results under several rapid irradiance variations proved that the proposed MPPT method fulfilled the highest efficiency comparing to a conventional single neural network.
Abstract: The approach of subset selection in polynomial
regression model building assumes that the chosen fixed full set of
predefined basis functions contains a subset that is sufficient to
describe the target relation sufficiently well. However, in most cases
the necessary set of basis functions is not known and needs to be
guessed – a potentially non-trivial (and long) trial and error process.
In our research we consider a potentially more efficient approach –
Adaptive Basis Function Construction (ABFC). It lets the model
building method itself construct the basis functions necessary for
creating a model of arbitrary complexity with adequate predictive
performance. However, there are two issues that to some extent
plague the methods of both the subset selection and the ABFC,
especially when working with relatively small data samples: the
selection bias and the selection instability. We try to correct these
issues by model post-evaluation using Cross-Validation and model
ensembling. To evaluate the proposed method, we empirically
compare it to ABFC methods without ensembling, to a widely used
method of subset selection, as well as to some other well-known
regression modeling methods, using publicly available data sets.
Abstract: Soil chemical and physical properties have important
roles in compartment of the environment and agricultural
sustainability and human health. The objectives of this research is
determination of spatial distribution patterns of Cd, Zn, K, pH, TNV,
organic material and electrical conductivity (EC) in agricultural soils
of Natanz region in Esfehan province. In this study geostatistic and
non-geostatistic methods were used for prediction of spatial
distribution of these parameters. 64 composite soils samples were
taken at 0-20 cm depth. The study area is located in south of
NATANZ agricultural lands with area of 21660 hectares. Spatial
distribution of Cd, Zn, K, pH, TNV, organic material and electrical
conductivity (EC) was determined using geostatistic and geographic
information system. Results showed that Cd, pH, TNV and K data
has normal distribution and Zn, OC and EC data had not normal
distribution. Kriging, Inverse Distance Weighting (IDW), Local
Polynomial Interpolation (LPI) and Redial Basis functions (RBF)
methods were used to interpolation. Trend analysis showed that
organic carbon in north-south and east to west did not have trend
while K and TNV had second degree trend. We used some error
measurements include, mean absolute error(MAE), mean squared
error (MSE) and mean biased error(MBE). Ordinary
kriging(exponential model), LPI(Local polynomial interpolation),
RBF(radial basis functions) and IDW methods have been chosen as
the best methods to interpolating of the soil parameters. Prediction
maps by disjunctive kriging was shown that in whole study area was
intensive shortage of organic matter and more than 63.4 percent of
study area had shortage of K amount.