The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON SYSTEMS
Transactions ID Number: 52-532
Full Name: Abbas Pourzaki
Position: Doctor (Researcher)
Age: ON
Sex: Male
Address: 1Department of Electrical and Electronic Engineering, Khorasan Research Institute for Food Science & Technology, Mashad, Iran.
Country: IRAN
Tel:
Tel prefix:
Fax:
E-mail address: pourzaki@krifst.ir
Other E-mails:
Title of the Paper: Study on Artificial Neural Networks Applications in Nonlinear Behavioral Modeling Patents
Authors as they appear in the Paper:
Email addresses of all the authors:
Number of paper pages: 18
Abstract: The level of the Complexity of the RF devices in communication systems has been steadily increased significantly, to support multiple standards, multiple frequency bands, need for higher bandwidth and stringent adjacent channel specifications.In this paper we have studied Artificial Neural Networks Applications in Nonlinear Behavioral Modeling Patents. From the recent patents, one may anticipate that behavioral models are proper tools to predict system performance without complexity of full simulation of the circuit as well as regarding their ability to reduce the computation time required for the analysis of the large circuits and systems. As a demonstration of the capability of that, new modeling methods are described, which combines electrical circuit theory and numerical analysis, e.g. neural networks error backpropagation or neural networks time domain extensions for dynamic systems. In this paper generating artificial neural network models for nonlinear de!
vices and/or circuits is described which is extracted from the frequency domain or time domain measured or simulated data. The ANN models provide accurate description for the device or circuit over the full span of the measurement or simulation. These techniques are very useful for neural-based computer-aided design and could be used for analytically unified dc, small signal and nonlinear device modeling. Various neural networks are used such as Radial Basis Function (RBF), Multilayer Perceptron (MLP) and Recurrent Neural Network (RNN). Finally, current technology has been discussed and future technology about this subject (can be omitted) is prospected.
Keywords: Behavioral modeling, neural network, perceptron, radial basis, nonlinear model, training, backpropagation
EXTENSION of the file: .doc
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress:
IP ADDRESS: 217.219.244.24
