Applications of Neural Networks in Electromagnetics

Christos Christodoulou and Michael Georgiopoulos

Artech House

2001
530 pages

Description

The high-speed capabilities and "learning" abilities of neural networks can be applied to quickly solving numerous complex optimization problems in electromagnetics, and this book shows you how. Even if you have no background in neural networks, this book helps you understand the basics of each main network architecture in use today, including its strengths and limitations. Moreover, it gives you the knowledge you need to identify situations when the use of neural networks is the best problem-solving option.

Organized in a modular format that lets you find and use the information you need fast, the book includes five full chapters that zero-in on specific neural network applications. Included are chapters on antennas, remote sensing and target classification, microwave devices and circuit modeling, and real-time performance systems for military and civilian systems, such as GPS and mobile communications. You also see how neural networks can be used in conjunction with other methods, such as the finite element method, the finite difference method, and the method of moments. More than 700 equations and over 200 illustrations are included, and MATLAB code for applications using neural network technology is found in relevant chapters.

Table of Contents

Preface

Acknowledgments

Introduction to Neural Networks

Preliminaries

Benefits of Neural Networks

Models of a Neuron

Types of Activation Function

Neural Network Architectures

Learning Procedures

Learning Tasks

Knowledge Representation

Brief History of Neural Networks

Why Neural Networks in Electromagnetics

Single-Layer and Multilayer Perceptron Networks

Introduction

The Single-Layer Perceptron

Perceptron Learning Algorithm

Adaline Network

Multilayer Perceptron

The Back-Propagation Algorithm

Issues With Back-Propagation Learning

Variations of the Back-Propagation Algorithm

The Mulitlayer Perceptron Neural Network for an Automatic Target Recognition Application

MATLAB Code

Radial Basis Function Networks-Kohonen Networks

Introduction

Preliminaries of Radial Basis Function Neural Networks

Learning Strategies With Radial Basis Function Neural Networks

A Radial Basis Function Neural Network Algorithm

A Radial Basis Function Neural Network Example

Comparison of Radial Basis Function Neural Network Learning Strategies

Issues With Radial Basis Function Neural Network Learning

The General Regression Neural Network (GRNN)

MATLAB Code

Adaptive Resonance Theory Neural Networks

Introduction

The Fuzzy ARTMAP Neural Network

Templates in Fuzzy ARTMAP: A Geometrical Interpretation

Example

Convergence Speed of Fuzzy ARTMAP

Order of Search in Fuzzy ARTMAP

Applications of Fuzzy ARTMAP

MATLAB Code

Recurrent Neural Networks

Introduction

Preliminaries of Associative Memories

The Hopfield Model

Associative Memory Application of the Hopfield Neural Network

Discussion

Optimization Problems Using the Hopfield Neural Network

A Problem in Communications Using the Hopfield Neural Network

The RTRL Neural Network

RTRL NN Examples

The Recurrent Time Recurrent Learning Neural Network for Channel Equalization

The Elman Neural Network

Elman Neural Network Examples

Angle of Arrival Estimation Using Elman Networks

MATLAB Code

Applications in Antennas

Introduction

Design of Gratings and Frequency Selective Surfaces

Neural Network-Based Adaptive Array Antennas

Beam Shaping With Antenna Arrays

Aperture Antenna Shape Prediction

Reflector Surface Error Compensation

Resonance Frequency of Triangular Microstrip Antennas

Design of Multilayer Phased Array Antennas

Applications in Radar and Remote Sensing

Introduction

Radar Target Classification

Classification of Radar Clutter

Remote Sensing

Applications in Mobile Communications

Introduction

Adaptive Antenna Array Processing

Neural Network-Based Direction Finding

Direction of Arrival for Multiple Sources Using Multilayer Neural Networks

Adaptive Nulling and Steering

Neural Network-Based Interference Cancellation

Applications in Microwave Circuits and Devices

Introduction

Simulation and Optimization of Microwave Devices and Circuits

Modeling of Passive Devices for MMIC Design

Speeding Up and Configuring the Optimum Size for a Neural Network

A Modular, Knowledge-Based Development of Libraries of Neural Network Models

Applications in Computational Electromagnetics

Introduction

Finite Element Applications

A General Neural Network Representation of FEM

A Neural Network Approach of the Method of Moments

Combination of the Piecewise Harmonic Balance Technique and Neural Networks