Advances in Information Technologies for Electromagnetics

Advances in Information Technologies for Electromagnetics Advances in Information Technologies for Electromagnetics Edited by Luciano Tarricone University of Lecce, Italy and Alessandra Esposito University of Lecce, Italy A C.I.P. Catalogue record for this book is available from the Library of Congress. ISBN-10 ISBN-13 ISBN-10 ISBN-13 1-4020-4748-7 (HB) 978-1-4020-4748-0 (HB) 1-4020-4749-5 (e-book) 978-1-4020-4749-5 (e-book) Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. www.springer.com Printed on acid-free paper All Rights Reserved © 2006 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed in the Netherlands Dedication This book is dedicated to Edoardo and Silvia Contents Contributing Authors Preface Acknowledgments xvii xxi xxvii 1 Parallel and Distributed Environments A. Esposito 1. 2. 3. INTRODUCTION BASIC CONCEPTS PARALLEL PROGRAMMING 3.1 Introduction 3.1.1 MPI 3.2 Performance Assessment DISTRIBUTED SYSTEMS 4.1 Introduction 4.2 RPC 4.3 Mobile Agent Framework THE WEB 5.1 XML 5.1.1 Introduction 5.1.2 XML Fundamentals 5.1.3 Namespaces 5.1.4 XML Schema 5.1.5 Applications 1 2 3 3 5 6 6 6 7 8 8 10 10 12 13 15 16 2 Object-Oriented Technologies A. Esposito 19 1. 2. INTRODUCTION OO PROGRAMMING 19 20 4. 5. 1 viii Contents 2.1 2.2 Basic Concepts Java 2.2.1 Introduction 2.2.2 The Language OO DISTRIBUTED FRAMEWORKS 3.1 Introduction 3.1.1 Java RMI 3.2 Java Mobile Agents 20 23 23 24 26 26 26 27 3 The Semantic Web A. Esposito 29 1. 2. INTRODUCTION DESCRIPTION LOGICS 2.1 Introduction 2.2 A Model for Reality: The TBox 2.2.1 Constructors 2.2.2 Axioms 2.3 The ABox 2.4 Reasoners TOOLS FOR THE SEMANTIC WEB 3.1 Languages 3.2 Reasoners 3.3 Tools for Building Ontologies 29 31 31 31 33 35 37 38 41 41 42 43 4 Web Services A. Esposito 45 1. 2. INTRODUCTION BASIC CONCEPTS 2.1 Web Services Architecture WEB SERVICES DESCRIPTION: WSDL AUTOMATIC DISCOVERY OF WEB SERVICES 4.1 UDDI 4.2 The Semantic Web Services 45 46 46 48 50 50 50 5 Grid Computing A. Esposito 55 1. 2. INTRODUCTION GC BASIC CONCEPTS 55 56 3. 3. 3. 4. Contents ix 3. THE GLOBUS TOOLKIT 3.1 GT and Web Services GT COMPONENTS JOB MANAGEMENT 5.1 GC for HPC INFORMATION SERVICES DATA MANAGEMENT 57 58 60 61 61 62 65 Complex Computational Electromagnetics using Hybridisation Techniques R. A. Abd-Alhameed and P. S. Excell 69 4. 5. 6. 7. 6 1. 2. 3. 4. 5. INTRODUCTION 70 1.1 Integral Equation Methods 70 70 1.2 Differential Equation Methods 71 1.3 The Advantages and Disadvantages of the Methods 72 1.4 Hybrid Methods 74 1.5 Literature Review OUTLINE OF THEORY AND IMPLEMENTATION OF HYBRID METHOD 80 2.1 Hybrid Treatment for Homogeneous Multiple Elements 80 2.1.1 Hybrid MoM/MoM Treatment for Two Elements (Sub-Matrices Iterative Technique) 81 2.1.2 Hybrid MoM/MoM Method for Two Elements 84 (Field Transfer Iterative Technique) 2.1.3 Extension of Hybrid MoM/MoM Method from Two Elements to Multiple Elements (Field Transfer Iterative Technique) 88 2.1.4 Hybrid MoM in Multiple Regions Using 91 the Equivalence Principle Surface INCIDENT WAVE EXCITATIONS IN THE FDTD METHOD 101 102 3.1 Total/Scattered Field Formulation in Three Dimensions MODIFIED TOTAL/SCATTERED FIELD FORMULATION FOR THE HYBRID TECHNIQUE 107 VALIDATION OF TOTAL/SCATTERED FIELD FORMULATION IMPLEMENTATION USING 110 HOMOGENEOUS FDTD IN MULTIPLE REGIONS x Contents 6. 112 113 116 120 123 HYBRID MOM/FDTD TECHNIQUE ALGORITHM 6.1 Theoretical Formulation 6.2 Multiple-Source Scattering Problems 7. NEC/FDTD HYBRID PROGRAM 8. FAR FIELD CALCULATIONS USING THE HYBRID CODE 9. NUMERICAL EXAMPLES USING THE HYBRID MoM/FDTD TECHNIQUE 10. SUMMARY 123 140 7 Enhanced EM software for Planar Circuits D. Vande Ginste, F. Olyslager, D. De Zutter and E. Michielssen 147 1. INTRODUCTION 1.1 Setting and Definition of the Research Topic 1.1.1 High-Frequency Applications and Design 1.1.2 Planar Circuits and Planar Solvers 1.1.3 Some Advantages and Drawbacks of BIE-MoM Based Planar Solvers 1.2 Methodology 1.2.1 Perfectly Matched Layer (PML) Based Green’s Functions 1.2.2 Iterative Solvers 1.2.3 Fast Multipole Method (FMM) 1.3 Outline CLASSICAL SOLUTION TECHNIQUE FOR MICROSTRIP STRUCTURES 2.1 Geometry of the Problem 2.2 The EFIE Description 2.3 The Green’s Dyadic G ee (r | r ') 2.3.1 Integral Representation 2.3.2 Sommerfeld-Integrals 2.4 The Method of Moments PERFECTLY MATCHED LAYER BASED GREEN’S FUNCTIONS FOR LAYERED MEDIA 3.1 The Perfectly Matched Layer Concept 3.1.1 The Split Field Formalism 3.1.2 Complex Coordinate Stretching Formalism 3.2 Closure of Open Microstrip Substrates 3.2.1 Procedure and Influence on the Green’s Functions 3.2.2 Complex Thickness 3.2.3 Dispersion Relations 148 148 148 149 2. 3. 150 151 151 153 154 155 156 156 157 158 158 160 161 163 163 163 164 165 165 166 167 Contents Series Expansion for the Green’s Dyadic G ee 3.3.1 Integral Representation 3.3.2 Gee , xx 3.3.3 Gee , xy 3.3.4 Closed-Form Expression for G ee 3.3.5 Important Remarks Concerning the Series Expansion A PML-MLMFA FOR THE MODELING OF LARGE PLANAR MICROSTRIP STRUCTURES 4.1 Introduction and Outline 4.2 Formulation of the Technique 4.2.1 The moment Matrix Written as Interactions Between Elementary Current Sources 4.2.2 Plane Wave Decomposition of the Hankel Function 4.2.3 Core Equation of the PML-MLFMA for Microstrip Structures 4.3 Implementation of the Technique 4.3.1 Construction of the MLFMA Tree 4.3.2 The Matrix-Vector Multiplication 4.4 Some Important Remarks about the Complexity of the PML-MLFMA 4.4.1 Memory and Computational Complexity 4.4.2 Mode Trimming l 4.4.3 Determination of the Sampling Rates 2QTX, n +1 4.5 Numerical Results 4.5.1 Validation of the Method 4.5.2 Computational and Memory Efficiency 4.5.3 Application Examples EXTENSIONS AND CONCLUSIONS 5.1 Extensions 5.1.1 Development of a Low-Frequency Algorithm 5.1.2 Combination of the HF- and the LF-Technique 5.1.3 Extension to General Multilayered Structures 5.2 Conclusions 3.3 4. 5. 8 1. 2. xi 168 168 168 172 172 173 174 174 175 175 176 178 180 180 185 195 195 195 196 197 197 204 206 210 210 210 213 214 215 Parallel Grid-enabled FDTD for the Characterization of Metamaterials L. Catarinucci, G. Monti, P. Palazzari and L. Tarricone 223 INTRODUCTION INTRODUCTION TO METAMATERIALS 2.1 DNG Metamaterials 223 224 225 Contents xii 3. 4. 5. 6. 7. NEGATIVE REFRACTION HOW TO SYNTHESIZE A DNG MEDIUM DNG MEDIA APPLICATIONS MODULATED SIGNALS IN A DNG MEDIUM 6.1 Dispersion 6.2 Gaussian Pulse in a DNG Slab NUMERICAL METHODS FOR METAMATERIALS 7.1 Bases for the FDTD Method 7.2 Parallel Grid-Enabled FDTD using MPI 7.3 Efficient Subgridding Technique for Parallel FDTD Algorithms: Variable Mesh FDTD 7.4 FDTD Methods and DNG Materials 7.5 DNG Slabs: Reflection by and Propagation in a DNG Slab 227 228 234 236 236 237 242 242 249 250 256 257 9 A Software Tool for Quasi-Optical Systems N. C. Albertsen, P. E. Frandsen and S. B. Sørensen 265 1. 2. INTRODUCTION REQUIREMENTS FOR QUASI-OPTICAL NETWORK DESIGN OUTLINE OF THE SOFTWARE SYSTEM ANALYSIS METHODS USER INTERFACE - THE FRAME EDITOR COMPONENTS AND OBJECTS: THE OBJECT WIZARD COMPLEX COMMANDS: THE COMMAND WIZARD FRAME CONNECTIONS AND 3D MODELLING EVALUATION AND FUTURE EXTENSIONS 265 3. 4. 5. 6. 7. 8. 9. 10 Cooperative Computer Aided Engineering of Antenna Arrays A. Esposito, L. Tarricone, L. Vallone and M. Vallone 1. 2. 3. 4. 5. INTRODUCTION CAE OF APERTURE ANTENNA ARRAYS GRID SERVICES AND SEMANTIC GRID SYSTEM ARCHITECTURE THE FRAMEWORK 5.1 Introduction 5.2 Grid Infrastructure 5.3 Encapsulation into Services 5.4 Ontology 5.4.1 Introduction 267 271 274 276 281 283 286 291 295 295 296 297 298 301 301 302 303 306 306 Contents 6. 5.4.2 Service Discovery 5.4.3 Service Orchestration 5.4.4 Service Binding 5.5 Client Application 5.5.1 Introduction 5.5.2 Service Discovery 5.5.3 Service Orchestration 5.5.4 Service Invocation CONCLUSIONS xiii 307 309 317 320 320 320 321 322 323 11 Distributed and Object-Oriented Computational Electromagnetics on the Grid D. Caromel, F. Huet, S. Lanteri and N. Parlavantzas 327 INTRODUCTION DISTRIBUTED OBJECTS: PROACTIVE 2.1 Basic Model 2.2 Mapping Active Objects to JVMs: Nodes 2.3 Deployment Descriptors 2.4 Group Communications OO DISTRIBUTED FINITE VOLUME SOLVER 3.1 Basic Architecture of the OO Model 3.2 Distribution and Parallelization BENCHMARKS 4.1 Comparison with a Fortran Implementation 4.2 Grid’5000 Experiments ON-GOING AND FUTURE WORK 5.1 Application Controlled Deployment 5.2 Enhancing Modifiability Through Components CONCLUSIONS 327 328 328 329 329 331 332 333 335 337 337 338 339 339 339 342 1. 2. 3. 4. 5. 6. 12 Software Agents for Parametric Computational Electromagnetics Applications D. G. Lymperopoulos, I. E. Foukarakis, A. I. Kostaridis, C. G. Biniaris and D. I. Kaklamani 1. 2. INTRODUCTION CLASSIFICATION OF PARAMETRIC PROBLEMS IN CEM 2.1 “Method-level” Parametric Analysis 2.2 “Application-level” Parametric Analysis 2.3 Population-Based Stochastic Optimisation 345 345 347 347 348 348 Contents xiv 3. 4. 5. MOBILE SOFTWARE AGENTS 349 3.1 The Mobile Agent Paradigm 349 351 3.2 Mobile Agents in CEM: The Master-Worker Model 3.2.1 The Master Agent 351 352 3.2.2 The Worker Agent 353 3.3 A Brief Comparison Between MAT and MPI or PVM A WEB-BASED MOBILE AGENT PLATFORM FOR PARAMETRIC CEM MODELING 355 4.1 Mobile Agent Platform Components 355 357 4.2 Communication Mechanisms 357 4.3 Web-Based Infrastructure 358 4.3.1 Interaction With the User 4.3.2 Servlets for Front/Back-End Communication 359 4.4 Conformal Array Modelling: A Modified Method of Auxiliary Sources (MMAS) Approach 360 4.4.1 Problem Formulation 360 362 4.4.2 Overview of the Model Geometry 4.4.3 Agent Deployment Mechanisms 363 364 4.4.4 Simulation Results 4.5 Electromagnetic Penetration Through Apertures: A Resonator Method of Moments (MoM) Model 365 4.5.1 Formulation of the Electromagnetic Problem 365 4.5.2 Parametric Simulations 369 4.5.3 Performance Results 369 INTRODUCING GENETIC SOFTWARE AGENTS 371 372 5.1 Distributed Genetic Algorithms with Agents 373 5.1.1 Entity Mappings 374 5.1.2 Parallel Processing Coordination 375 5.2 Proposed Architecture 375 5.2.1 Centralised Model 376 5.2.2 Decentralised Model 5.2.3 Hybrid Implementations 376 377 5.3 Conclusions 13 Web Services Enhanced Platform for Distributed Signal Processing in Electromagnetics I. E. Foukarakis, D. B. Logothetis, A. I. Kostaridis, D. G. Lymperopoulos and D. I. Kaklamani 1. 2. INTRODUCTION WEB SERVICES IN DISTRIBUTED SAR MODELLING AND SIGNAL PROCESSING 2.1 Platform Architecture 381 381 382 382 Contents Server Services 2.2.1 Node Management Service 2.2.2 Input Provider Service 2.2.3 Output Receiver Service 2.2.4 Scheduler 2.3 Node Services 2.3.1 Resource Manager Service 2.3.2 Task Executing Service 2.3.3 Remote Input Service 2.4 Other Issues 2.5 Imaging Radar Signal Processing 2.6 The Simulation Mechanism 2.7 Results and Conclusions 385 385 387 387 387 389 389 389 390 390 390 392 394 Grid-Enabled Transmission Line Matrix (TLM) Modelling of Electromagnetic Structures P. Russer, B. Biscontini and P. Lorenz 399 2.2 14 1. 2. 3. 4. 5. 6. 7. xv INTRODUCTION THE 3D-TLM METHOD MODELLING OF DIELECTRIC MEDIA PARALLELIZATION OF THE TLM METHOD 4.1 Domain Decomposition 4.2 Decomposition of the TLM Algorithm TLM-G: GRID-ENABLED TIME DOMAIN TRANSMISSION LINE MATRIX SYSTEM 5.1 The Components of the TLM-G System 5.2 The Relation Between YATWAD, YATD and the Components of the Globus Toolkit in the TLM-G System ANALYSIS OF THE PERFORMANCE OF THE TLM-G SYSTEM AND EXAMPLES 6.1 The Electromagnetic Performance of the TLM-G System 6.2 A Bowtie Antenna in a TLM-G System THE CIRCULAR CYLINDRICAL CAVITY RESONATOR 399 400 411 415 415 417 Glossary 433 Index 451 422 423 424 425 425 426 428 Contributing Authors Raed A. Abd-Alhameed University of Bradford, UK Niels Christian Albertsen Informatics and Mathematical Modelling, Technical University of Denmark Christos G. Biniaris School of Electrical and Computer Engineering, National Technical University of Athens, Greece Bruno Biscontini Technische Universität München, Munich, Germany Denis Caromel INRIA, France Luca Catarinucci University of Lecce, Italy Daniel De Zutter Ghent University, Belgium Alessandra Esposito University of Lecce, Italy Peter Stuart Excell University of Bradford, UK xvii xviii Contributing Authors Ioannis E. Foukarakis School of Electrical and Computer Engineering, National Technical University of Athens, Greece Poul Eric Frandsen TICRA Engineering Consultants, Copenhagen, Danmark Fabrice Huet INRIA, France Dimitra I. Kaklamani School of Electrical and Computer Engineering, National Technical University of Athens, Greece Antonis I. Kostaridis School of Electrical and Computer Engineering, National Technical University of Athens, Greece Stefane Lanteri INRIA, France Dyonisios B. Logothetis School of Electrical and Computer Engineering, National Technical University of Athens, Greece Petr Lorentz Technische Universität München, Munich, Germany Dimitrios G. Lymperopoulos School of Electrical and Computer Engineering, National Technical University of Athens, Greece Eric Michielssen University of Illinois at Urbana-Champaign, USA Giuseppina Monti University of Lecce, Italy Frank Olyslager Ghent University, Belgium Contributing Authors Paolo Palazzari ENEA-HPCN, Italy Nikos Parlavantzas INRIA, France Peter Russer Technische Universität München, Munich, Germany Stig Busk Sørensen TICRA Engineering Consultants, Copenhagen, Danmark Luciano Tarricone University of Lecce, Italy Laura Vallone University of Lecce, Italy Mariangela Vallone University of Lecce, Italy Dries Vande Ginste Ghent University, Belgium xix Preface The way of designing, tuning, trimming and realizing electromagnetic (EM) circuits and antennas has deeply changed in the last decades. The continuous growth in the variety of applications, services, and in the required performance, the reduction of production times, in other words the increasing complexity of the task microwave (MW) engineers are typically involved in, has leaded to a systematic use of computers and numerical methods in daily research and development workflow. Meanwhile, an extraordinary progress in Information and Communication Technologies (ICT) is a constant matter of fact in the present years, with a consequent availability of more and more powerful computing platforms and systems, or software methodologies, which are modifying the way we conceive the activities of Computer Aided Design (CAD) or Engineering (CAE) in any scientific area, and the EM context makes no exceptions. As a consequence, more and more attention is paid, even from the educational point of view, in the EM community, to ICT evolution, as it is rather evident the relevant impact that it can have on the efficiency, effectiveness and quality of EM devices. The very recent past has seen, in the ICT area, the affirmation of some new and important key-words. Apart from the most common (and abused), such as Internet-related ones, we want here to recall the concept of “system integration”, intended as the capability of generating added-value products by assembling together pre-existing tools, as well as the one of “distributed” and “cooperative” computing, intended as the capability of exploiting in a concerted manner computational or software resources not physically concentrated in one single site or platform. Inside these concepts, very wide and multi-folded, many other xxi xxii Preface focal issues can be discovered, representing in many cases hot cutting-edge points, such as the ones associated to software interoperability and reusability, high performance and grid computing, resource and process sharing, etc. The above enumerated key-words are more than “simple” technological or scientific turn-points, as they intimately affect the way people work and plan their daily activities (for instance, the more and more spread adoption of outsourcing processes, or the huge number of cooperative projects on a geographical scale). On the other hand, the EM community is, still today, far from being completely aware of the impressive potential impact that recent ICT progresses can have on itself. This is partly due to a physiological delay in the assimilation of complex and “foreign” approaches, and partly to a persistent lack of bridging tools between the EM world and the ICT one. This book is intended to bridge this gap, so that, on one side, EM researchers can identify new and promising ICT tools (already available, or to be consolidated in the immediate future), expected to significantly improve their daily EM investigation. On the other side, ICT experts can find here appealing ideas and scientific areas, perhaps not explored or considered till now, where ICT can play a major and innovative role. Consequently, EM researchers can find in this book interesting answers from ICT to some questions emerging often and often, for example how to get more computing power, or how to improve the capability to manage complex projects with special software solutions (Chapters 6 to 9). Furthermore, they can also see sample applications where suitable EM theoretical formulations, or algorithms, render EM numerical methods more amenable to take advantage from new ICT tools (Chapters 10 to 14). The now enumerated chapters, proposing real applications, are preceded by an introductory part (Chapters 1 to 5), containing foundations to Information Technology, so that even beginners in the ICT area are introduced to basic concepts playing a key-role in further parts of the book. More in detail, the book is structured as follows. Chapter 1 provides a very simple taxonomy of terms and an introduction to basic concepts of parallel and distributed computing. As for parallel computing, a special focus is devoted to MPI, for its recent affirmation as standard. The importance of MPI is confirmed by its wide use in grid computing environments, as demonstrated in Chapter 5 and in Chapter 8. As for distributed computing, Chapter 1 deals with two technologies, the low level technique named RPC and the powerful mobile agents paradigm. The former is a well known technique widely adopted in the distributing computing arena, as confirmed by the availability of several libraries for its implementation. Among them, we recall RMI and XML-RPC which are Preface xxiii widely used throughout the book (Chapters 10, 11, 12, 13). The latter (mobile agents) is the core of Chapter 11 and is one of the most promising technologies of the moment. Finally, Chapter 1 introduces the basic concepts of XML, the meta-language for exchanging information through the Web, which provides the foundations for Web Services and the last generation of Grid Computing frameworks (including Semantic Grids). An EM application adopting such technologies is proposed in Chapter 10. Other relevant technologies related to distributed computing are dealt with in Chapter 2, for their relationship with object oriented (OO) concepts, and in Chapter 4 and 5, where the two hot technologies of Web Services and Grid Computing are addressed respectively. More in detail, Chapter 2 is devoted to the introduction of basic concepts and terms of object oriented programming and software designing model. The chapter focuses on the Web-oriented OO language par excellence: Java. A description of the language is provided together with an overview of Java-based object oriented distributed frameworks, i.e. Java RMI and Java mobile agents. The former is widely used in Chapter 11, the latter is adopted in Chapter 12 and 13. Chapter 3 overviews the basic concepts behind the Semantic Web (and consequently the Semantic Grids). The Semantic Web encodes knowledge by means of an appropriate language, so that electronic agents search information on the bases of human-readable queries. In this way, discovery and choreography of resources is automated and made easier also in complex environments, such as Internet-wide grid frameworks. As shown in Chapter 10, where a practical EM application benefiting from Semantic Web technologies is described, this can give a huge momentum to the solution of complex multidisciplinary problems. Chapter 4 introduces the concepts behind Web Services (WS) and the main standards supporting them. WS propose a new model for implementing applications, thus promoting reusability and cooperation. An example of application of WS to EM is provided in Chapter 13, whilst Chapter 10 shows the potentials of integrating WS with Grid technologies. Finally, Chapter 5 shows how grid computing has recently embraced WS concepts, driving a process of fusion of WS and grids. In this way, sharing of standard hardware and software computing resources is joined with the potential of orchestrating autonomously developed software components offered by WS technologies. Examples of these capabilities are provided in Chapter 8, 10 and 14. Chapter 6 opens the part of the book devoted to EM applications. In chapter 6, the relevant issue of hybrid numerical methods is addressed. A wide review of EM numerical methods is proposed, so that the importance of method hybridization is cleared. Sample applications are proposed, and xxiv Preface crucial topics for effective hybridizations are focused, such as the one of domain partitioning. Chapter 7 proposes an interesting approach where the use of suitable EM theoretical formulations, and algorithmic solutions, is the right way to reduce numerical complexity. The use of Perfectly Matched Layers (PML) leads to a series representation for Green’s functions of planar circuits. The terms in this series allow for the application of a Multilevel Fast Multipole Algorithm (MLFMA) for the analysis of very large planar structures, or of small circuits with high geometric detail, with impressive computational performance. Chapter 8 is an example of how a standard numerical method (FDTD) can be implemented so to take advantage of parallel computing platforms, even with very low costs (consider for instance computational grids). Furthermore, suitable algorithmic solutions allow to couple such advantage with the selection of variable meshes. The resulting parallel grid-enabled variable-mesh FDTD tool can play a major role in the investigation of the hot topic of metamaterial properties, with a special focus on finitebandwidth signal propagation inside double-negative materials. Chapter 9 proposes a software tool for designing and analyzing quasioptical complex systems. Along with a general discussion of the analysis methods available, particular emphasis is put on the user interface and other relevant software components, demonstrating how an advanced use of graphical facilities can turn into an impressive added-value generator. Chapter 10 is devoted to the use of Grid Computing, and more specifically to Semantic Grid, as a tool for cooperative computer-aided engineering (CAE) of antenna arrays. The CAE of aperture antenna arrays is a good example of a complex application, merging different skills and scientific knowledge, with a consequent potential demand for cooperation among several research groups. The adoption of Semantic Grids is the right answer to this demand. It also allows an implementation of the CAE environment in a service oriented framework, where CAE components are encapsulated in grid services and exploitable remotely through the grid. The chapter also profiles the attractiveness of ontologies as an immediate and future turning point. Chapter 11 is one more evidence of the relevance grid computing is gaining in the EM community. More specifically, this chapter suggests an interesting coupling between grid technologies and object oriented methodologies, with the final goal of developing high performance numerical methods for the solution of systems of PDEs (Partial Differential Equations). An open source middleware for the grid, featuring distributed objects and components, is proposed to design and implement an object- Preface xxv oriented time domain finite volume solver on unstructured meshes for the 3D modelling of EM propagation. Chapter 12 proposes the application of novel networking software technologies in distributed parallel CEM computing. Web services and mobile agents are used to solve demanding parametric CEM problems, this resulting in platform-independent concurrent computing solutions. Conformal array modelling with the method of Auxiliary Sources, and the study of EM penetration through aperture with the Method of Moments are proposed as testbeds. In addition, Genetic Software Agents are introduced, i.e. mobile agent entities with the ability to carry out Genetic Search Optimisations in a collaborative scheme. Their perspective use is outlined. On the basis of the previous chapter, Chapter 13 extends the concepts of Web services to develop an enhanced distributed platform, discussing also architectural issues. The resulting application is service-oriented and the implementation is based on the Simple Object Access Protocol specifications. The platform is tested with a problem of microwave imaging using a coherent Synthetic Aperture Radar (SAR) sensor. Finally, Chapter 14 proposes the use of grid computing for the highperformance implementation of the Transmission Line Matrix (TLM) method. The parallelization of the TLM algorithm is performed by segmentation of the TLM state vector, whilst system identification and spectral analysis approaches allow a considerable reduction of numerical effort. In conclusion, the book suggests some ICT concepts, destined to play a major role in the current and future EM context. These concepts are introduced, at a beginner level, in the former part of the work. Their effective impact on real EM problems is then outlined in the latter part of the book, with a wide variety of applications. Acknowledgments The editors thank so much Barbara Pici for her precious editing work and Laura Vallone for her contribution to a global revision of the book. xxvii