Search Results: 1 - 10 of 100 matches for " "
All listed articles are free for downloading (OA Articles)
Page 1 /100
Display every page Item
Novel LDPC Decoder via MLP Neural Networks  [PDF]
Alireza Karami,Mahmoud Ahmadian Attari
Mathematics , 2014,
Abstract: In this paper, a new method for decoding Low Density Parity Check (LDPC) codes, based on Multi-Layer Perceptron (MLP) neural networks is proposed. Due to the fact that in neural networks all procedures are processed in parallel, this method can be considered as a viable alternative to Message Passing Algorithm (MPA), with high computational complexity. Our proposed algorithm runs with soft criterion and concurrently does not use probabilistic quantities to decide what the estimated codeword is. Although the neural decoder performance is close to the error performance of Sum Product Algorithm (SPA), it is comparatively less complex. Therefore, the proposed decoder emerges as a new infrastructure for decoding LDPC codes.
Performance Analysis of Various Activation Functions in Generalized MLP Architectures of Neural Networks  [PDF]
Bekir Karlik and A. Vehbi Olgac
International Journal of Artificial Intelligence and Expert Systems , 2010,
Abstract: The activation function used to transform the activation level of a unit (neuron) into an outputsignal. There are a number of common activation functions in use with artificial neural networks(ANN). The most common choice of activation functions for multi layered perceptron (MLP) isused as transfer functions in research and engineering. Among the reasons for this popularity areits boundedness in the unit interval, the function’s and its derivative’s fast computability, and anumber of amenable mathematical properties in the realm of approximation theory. However,considering the huge variety of problem domains MLP is applied in, it is intriguing to suspect thatspecific problems call for single or a set of specific activation functions. The aim of this study is toanalyze the performance of generalized MLP architectures which has back-propagation algorithmusing various different activation functions for the neurons of hidden and output layers. Forexperimental comparisons, Bi-polar sigmoid, Uni-polar sigmoid, Tanh, Conic Section, and RadialBases Function (RBF) were used.
P. Kala
Journal of Global Research in Computer Science , 2012,
Abstract: This paper presents the designing of a rectangular micro-strip antenna using Multilayer Preceptron (MLP) neural network model. The model was trained for 40 set of input-output (length of patch-resonant frequency) parameters. The output error and time delay (no. of epochs) were optimized by changing the learning rate and momentum term. In case of testing the output of the present model (resonant frequency) is found in good agreement with theoretical results.
Applying Neural Network Architecture for Inverse Kinematics Problem in Robotics  [PDF]
Bassam Daya, Shadi Khawandi, Mohamed Akoum
Journal of Software Engineering and Applications (JSEA) , 2010, DOI: 10.4236/jsea.2010.33028
Abstract: One of the most important problems in robot kinematics and control is, finding the solution of Inverse Kinematics. Inverse kinematics computation has been one of the main problems in robotics research. As the Complexity of robot increases, obtaining the inverse kinematics is difficult and computationally expensive. Traditional methods such as geometric, iterative and algebraic are inadequate if the joint structure of the manipulator is more complex. As alternative approaches, neural networks and optimal search methods have been widely used for inverse kinematics modeling and control in robotics This paper proposes neural network architecture that consists of 6 sub-neural networks to solve the inverse kinematics problem for robotics manipulators with 2 or higher degrees of freedom. The neural networks utilized are multi-layered perceptron (MLP) with a back-propagation training algorithm. This approach will reduce the complexity of the algorithm and calculation (matrix inversion) faced when using the Inverse Geometric Models implementation (IGM) in robotics. The obtained results are presented and analyzed in order to prove the efficiency of the proposed approach.
Inverse problems and invisibility cloaking for FEM models and resistor networks  [PDF]
Matti Lassas,Mikko Salo,Leo Tzou
Mathematics , 2013,
Abstract: In this paper we consider inverse problems for resistor networks and for models obtained via the Finite Element Method (FEM) for the conductivity equation. These correspond to discrete versions of the inverse conductivity problem of Calder\'on. We characterize FEM models corresponding to a given triangulation of the domain that are equivalent to certain resistor networks, and apply the results to study nonuniqueness of the discrete inverse problem. It turns out that the degree of nonuniqueness for the discrete problem is larger than the one for the partial differential equation. We also study invisibility cloaking for FEM models, and show how an arbitrary body can be surrounded with a layer so that the cloaked body has the same boundary measurements as a given background medium.
Physical Modeling and Control of Electric Arc Furnace by Neural Inverse Model
M.M. Lafifi,B. Boulebtateche,S. Afifi,Z. Oualia,R. Lakel,M. Bedda,F. Labed
Asian Journal of Information Technology , 2012,
Abstract: In this study, we try to answer a certain number of questions raised in the electric arc furnace control. In this study, we mention the studied choice of the physical model which approaches best the Electric Arc Furnace (E.A.F). The nonlinear model uses the conductance like electric parameter of adjustment, whereas usually one uses the resistance of arc as parameter of adjustment although its accessibility remains implicit. In order to answer the preset objectives, i.e. to control the process of the E.A.F. and this with an acceptable stability in the respect of the constraints and minimization of energy, it is proposed a structure of control by neural inverse model implemented by a Multi-Layer Perceptron network (MLP). The technique of training data base uses the retro propagation of error gradient algorithm. An application on MATLAB SIMULINK made it possible to simulate this structure; convincing results are recorded.
FEM and ANN combined approach for predicting pressure source parameters at Etna volcano
A. Di Stefano, G. Currenti, C. Del Negro, L. Fortuna,G. Nunnari
Nonlinear Processes in Geophysics (NPG) , 2010,
Abstract: A hybrid approach for forward and inverse geophysical modeling, based on Artificial Neural Networks (ANN) and Finite Element Method (FEM), is proposed in order to properly identify the parameters of volcanic pressure sources from geophysical observations at ground surface. The neural network is trained and tested with a set of patterns obtained by the solutions of numerical models based on FEM. The geophysical changes caused by magmatic pressure sources were computed developing a 3-D FEM model with the aim to include the effects of topography and medium heterogeneities at Etna volcano. ANNs are used to interpolate the complex non linear relation between geophysical observations and source parameters both for forward and inverse modeling. The results show that the combination of neural networks and FEM is a powerful tool for a straightforward and accurate estimation of source parameters in volcanic regions.
Implementation of Artificial Neural Network applied for the solution of inverse kinematics of 2-link serial chain manipulator.
Satish Kumar,Kashif Irshad
International Journal of Engineering Science and Technology , 2012,
Abstract: In this study, a method of artificial neural network applied for the solution of inverse kinematics of 2-link serial chain manipulator. The method is multilayer perceptrons neural network has applied. This unsupervised method learns the functional relationship between input (Cartesian) space and output (joint) space based on a localized adaptation of the mapping, by using the manipulator itself under joint control and adapting the solution based on a comparison between the resulting locations of the manipulator's end effectors in Cartesian space with the desired location. Even when a manipulator is not available; the approach is still valid if the forward kinematic equations are used as a model of the manipulator. The forward kinematic equations always have a unique solution, and the resulting Neural net can be used as a starting point for further refinement when the manipulator does become available. Artificial neural network especially MLP are used to learn the forward and the inverse kinematic equations of two degrees freedom robot arm. A set of some data sets were first generated as per the formula equation for this the input parameter X and Y coordinates in inches. Using these data sets was basis for the training and evaluation or testing the MLP model. Out of the sets data points, maximum were used as training data and some were used for testing for MLP. Backpropagation algorithm was used for training the network and for updating the desired weights. In this work epoch based training method was applied.
Alternative Sensor System and MLP Neural Network for Vehicle Pedal Activity Estimation  [PDF]
Ahmed M. Wefky,Felipe Espinosa,José A. Jiménez,Enrique Santiso,José M. Rodríguez,Alfredo J. Fernández
Sensors , 2010, DOI: 10.3390/s100403798
Abstract: It is accepted that the activity of the vehicle pedals (i.e., throttle, brake, clutch) reflects the driver’s behavior, which is at least partially related to the fuel consumption and vehicle pollutant emissions. This paper presents a solution to estimate the driver activity regardless of the type, model, and year of fabrication of the vehicle. The solution is based on an alternative sensor system (regime engine, vehicle speed, frontal inclination and linear acceleration) that reflects the activity of the pedals in an indirect way, to estimate that activity by means of a multilayer perceptron neural network with a single hidden layer.
Modeling Radiated Electromagnetic Emissions of Electric Motorcycles in Terms of Driving Profile Using Mlp Neural Networks
Ahmed Wefky;Felipe Espinosa;Luis de Santiago;Alfredo Gardel;Pedro Revenga;Miguel Martinez
PIER , 2013, DOI: 10.2528/PIER12102510
Abstract: Current automotive electromagnetic compatibility (EMC) standards do not discuss the effect of the driving profile on real traffic vehicular radiated emissions. This paper describes a modeling methodology to evaluate the radiated electromagnetic emissions of electric motorcycles in terms of the driving profile signals such as the vehicle velocity remotely controlled by means of a CAN bus. A time domain EMI measurement system has been used to measure the temporal evolution of the radiated emissions in a semi-anechoic chamber. The CAN bus noise has been reduced by means of adaptive frequency domain cancellation techniques. Experimental results demonstrate that there is a temporal relationship between the motorcycle velocity and the radiated emission power in some specific frequency ranges. A Multilayer Perceptron (MLP) neural model has been developed to estimate the radiated emissions power in terms of the motorcycle velocity. Details of the training and testing of the developed neural estimator are described.
Page 1 /100
Display every page Item

Copyright © 2008-2017 Open Access Library. All rights reserved.