WELCOME to the website of

Adrianus T. de Hoop

Lorentz Chair Emeritus Professor

Laboratory of Electromagnetic Research
Faculty of Electrical Engineering, Mathematics and Computer Sciences
Delft University of Technology
Mekelweg 4 • 2628 CD Delft • the Netherlands
E: a.t.dehoop@tudelft.nl
T: +31 (0)10 5220049 (home)
W: www.atdehoop.com

Short biography

Adrianus Teunis de Hoop was born in Rotterdam, the Netherlands, on 24 December 1927. He received his MSc-degree in Electrical Engineering (1950) and his PhD-degree in the Technological Sciences (1958) from Delft University of Technology, Delft, the Netherlands, both with the highest distinction (cum laude). He served Delft University of Technology as an Assistant Professor (1950-1957), Associate Professor (1957-1960) and Full Professor in Electromagnetic Theory and Applied Mathematics (1960-1996). Since 1996 he is Lorentz Chair Emeritus Professor in the Faculty of Electrical Engineering, Mathematics Computer Sciences of this University. In 1970 he founded at Delft the Laboratory of Electromagnetic Research, which has developed into a world-class center for electromagnetics, having a huge impact on the world's electromagnetic community and on electromagnetic research and education in the Netherlands.

Dr. De Hoop's research interests are in the broad area of wavefield modeling in acoustics, electromagnetics and elastodynamics. His interdisciplinary insights and methods in this field can be found in his seminal Handbook of Radiation and Scattering of Waves (1995) [Electronic reproduction 2008 (with corrections), freely downloadable, for private use, in .pdf], with wavefield reciprocity serving as one of the unifying principles governing direct and inverse scattering problems and wave propagation in complex (anisotropic and dispersive) media. He spent a year (1956-1957) as a Research Assistant with the Institute of Geophysics, University of California at Los Angeles, CA, USA, where he pioneered a modification of the Cagniard technique for calculating impulsive wave propagation in layered media, later to be known as the "Cagniard-DeHoop technique". This technique is presently considered as a benchmark tool in analyzing time-domain wave propagation. During a sabbatical leave at Philips Research Laboratories, Eindhoven, the Netherlands (1976-1977), he was involved in research on magnetic recording theory.

Since 1982, Dr. De Hoop is, on a regular basis, Visiting Scientist with Schlumberger-Doll Research, formerly at Ridgefield, CT, now at Cambridge, MA, USA, where he contributes to research on geophysical applications of acoustic, electromagnetic and elastodynamic waves. Grants from the "Stichting Fund for Science, Technology and Research" (founded by Schlumberger Limited) supported his research at Delft University of Technology. He was awarded the 1989 Research Medal of the Royal Institute of Engineers in the Netherlands, the IEEE 2001 Heinrich Hertz Gold Research Medal, the 2002 URSI (International Scientific Radio Union) Balthasar van der Pol Gold Research Medal and in 2014 the Honorary Membership of the Society of Exploration Geophysicists (SEG), Tulsa OK, USA. [Citation]

In 2003, H.M. the Queen of the Netherlands appointed him "Knight in the Order of the Netherlands Lion".

Dr. De Hoop is a Member of the Royal Netherlands Academy of Arts and Sciences and a Foreign Member of the Royal Flemish Academy of Belgium for Science and Arts. He holds an Honorary Doctorate (1981) in the Applied Sciences from Ghent University, Belgium and an Honorary Doctorate (2008) in the Mathematical, Physical and Engineering Sciences from Växjö University (since 2010, Linnaeus University), Växjö, Sweden. [Picture]

Recently, he is exploring a method for computing pulsed electromagnetic fields in strongly heterogeneous media with application to (micro- or nano-scale) integrated circuits and a methodology for time-domain pulsed-field antenna analysis, design and optimization for mobile communication and radar applications.

His avocation is playing the piano (and in the past, performing choral music with the Rotterdam Philharmonic Choir).

Main research interests

The author’s main research interests are in the broad area of the analytical and computational modeling of wave and diffusive fields in acoustics, electromagnetics and elastodynamics in complex (anisotropic and dispersive) media, with their wealth of applications in physics and engineering. As is emphasized in his Handbook of Radiation and Scattering of Waves (1995), [Electronic reproduction 2008 (with corrections), freely downloadable, for private use, in .pdf], they share a number of unifying principles, such as the tensorial structure of their governing field equations and the reciprocity properties of the time-convolution and the time-correlation type of their solutions. The space-time structure of the fundamental equations entails that the physics of the pertaining phenomena does evolve in space-time. From this perspective, frequency-domain descriptions of the pertaining phenomena can be classified as "mathematical artifacts". For this reason, the author invariably focuses on time-domain methods and techniques. Another reason for avoiding the frequency domain as a method of analysis is the following. One of the basic prerequisites for a mathematical uniqueness proof of any of the constructed solutions is the property of causality. Now, for the real-valued frequency parameter occurring in the time Fourier transformation that links the time domain to the frequency domain, this property is lost. On account of this, other artifacts have to be called for to ensure the uniqueness of the solution of the (frequency-domain counterparts of the) problems at hand.

With the view on recent developments in inter- and intra-device/system electromagnetic radiative digital signal transfer in microwave electronics, a study has been performed towards generalizations of the Cagniard-DeHoop technique that provide analytical tools for judging the signal integrity in the pertaining transfers. In this respect also the demand for strict time-domain regulations for pulsed-field ElectroMagnetic Interference (EMI) invokes analytical time-domain studies in wave field transfer.

Computational field discretization methods (general aspects)

Any computational method for solving wave or diffusion problems of interest to configurations met in physical or engineering practice amounts to discretizing the field quantities in space-time in conjunction with a discretization of the physical properties of the media in which the field quantities occur. Standard procedure for the discretization of the constitutive medium properties is to assign piecewise constant values to them. As a consequence, jump discontinuities in parameter values and, correspondingly, jump discontinuities in (certain components of) the field quantities occur. As a consequence of this, the property of differentiability of the field quantities is lost and the field equations in their form of (partial) differential equations are excluded to serve as the point of departure. Now, for all field and wave problems in acoustics, electromagnetics and elastodynamics related equivalent integral formulations exist that can be organized such that in them only field components that are continuous across the surfaces of jump discontinuity in constitutive properties do occur. These space-time field integral relations are the ones to serve as the starting point for an appropriate computational procedure. From the point of view of algebraic topology, the simplex is the fundamental geometrical building block and is appropriate for the purpose. Since in computational modeling the majority of configurations is taken to be time invariant, space-time can be taken as cylindrical in the time direction. Interpolation (for example, linear interpolation) on the thus constructed grid yields a representation of the computed field throughout the domain of computation.

Perfectly matched 3D Cartesian coordinates embedding

In typical practical (3D) configurations memory size and computation time force the actual computation to be restricted to a domain of interest that is as small as possible. To guarantee that the boundaries of this domain of interest do not, up to the desired accuracy, deteriorate the problem solution, the domain is padded with a so-called perfectly matched embedding. Such an embedding can be designed as to mimick the reflectionless absorption of the field generated in the domain of interest to any desired degree and thus reducing the spurious reflection arising from the termination of the domain of computation to the level of acceptance. The time-dependent, causality preserving, 3D Cartesian coordinate stretching procedure provides the tool for this.

The periodic boundary condition

The domain of computation, that consists of the domain of interest as it is padded with the time-dependent, causality preserving, 3D Cartesian coordinate stretched perfectly matched embedding has to be terminated with uniqueness conditions satisfying boundary conditions. The periodic boundary condition satisfies this requirements and is to be preferred over boundary conditions that force the normal component of the area density of field power flow on the boundary to be zero, because of the property that the spurious reflections arising from the truncation of the domain of computation are uniformly distributed over this domain irrespective of the location of the sources that generate the field

Analytical benchmark configurations - Cagniard-DeHoop method

To check a number of aspects related to the consistency and the accuracy of the computational procedure at hand, certain benchmark problems are needed, whose solutions are constructed with the aid of analytical techniques. In this respect, a collection of radiation, propagation, reflection and transmission problems that are solvable with the modified Cagniard technique (’Cagniard-De Hoop method’) can serve the purpose. An essential tool in this technique is the unilateral time Laplace transformation with real, positive transform parameter that has been introduced by L. Cagniard (1939) in relation to a problem in impulsive seismic wave propagation and that has been simplified by A.T.de Hoop (1960).

Error criterion for the mismatch in the field equations

Substitution of the (interpolated) computational field in the exact, physical field equations yields a distribution of source terms throughout the domain of computations that differ from their prescribed values. Some positive definite measure for the action of these spurious source distributions would provide a quantitative measure for the accuracy of the obtained solution. Once such a, preferably physics-based, error criterion would have been constructed, it could possibly also be the basis for formulating the discretization procedure at hand as an optimization procedure, with the corresponding iterative decrease of the norm of the mismatch procedure that standardly finds application in linear field source and parameter inversion problems.

Modern tensorial electromagnetic field theory

A recent investigation has revealed that a substantial gain in efficiency in teaching electromagnetic theory and handling electromagnetic field problems can be achieved by abandoning the standard vector characterization of the field and source quantities and introducing a tensor approach where the electric field and source quantities are defined as tensors of rank one and the magnetic field and source quantities as anti-symmetric tensors of rank two. Using the subscript notation and the Einstein summation convention (summation over repeated subscripts in a product of two tensors), all operations reduce to elementary ones. It also proves advantageous to consider spacetime as an affine space where the time coordinate in the temporal constituent is real variable that has no specific connection with the position coordinates in the spatial constituent (with a Euclidean metric), rather than a metric space with a (non-definite) Lorentz metric. Through this formalism, it can be shown that the theory of special relativity (covariance of the structural form of the field equations in the space and time coordinates handled by two observers in uniform, rectilinear relative motion) can be generalized to spatial dimensions. The pertaining (generalized) Lorentz transformations lead to the surprising result that the volume density of magnetic occurring in it can be conceived as to consist of moving particles with a magnetic charge that is a completely anti-symmetric tensor of rank three. The implications of this are presently under study. [ Article in .pdf format]

Array-structured wavefield physics in affine (N+1)-spacetime

Presently, the author's research is focused on developing, in accordance with Percy William Bridgman's operational approach [1], an axiomatic, array-structured wavefield physics theory in (algebraic) affine spacetime as put forward by Hermann Weyl in his book Space-Time-Matter [2], the latter as an alternative to Albert Einstein's (geometrical) tensor theory [3]. The research aims at accommodating all transport properties of waves as the carriers of (physical) information, energy and momentum in their capability of enabling the interaction of data and ideas between the members of the 'peer group of physicists'.
Of particular importance is the question whether this approach can shed light on the interaction between waves and (elementary) particles, a problem that Dirac expressed himself as 'having been unable to solve' [4].
[1] Bridgman, P.W., The Nature of Physical Theory,
Princeton University Press, Princeton NJ, 1936.

[2] Weyl, H., Space-Time-Matter,
Dover Publications, Mineola NY, 1952.

[3] Einstein, A., The Meaning of Relativity,
Princeton University Press, Princeton NJ, 1955.

[4] Farmelo, G., The Strangest Man: The hidden Life of Paul Dirac, Quantum Genius,
Faber and Faber, London, 2009, p.409.

Curriculum vitae

Personal data
Professional Career
Honors and Awards
Personal Research Grants
List of PhD-graduates

List of Publications

List of books
List of papers
List of conference proceedings
List of conference abstracts

Miscellaneous presentations


Course material

Course material