Why would anyone write yet another book about radar imaging? Although the field is relatively specialised and little known to the general public, there is already a lot of high quality literature available. We in fact have excellent reasons to do so.
The first is that the situation has changed due to new techniques and recent projects: the availability of data from new and efficient space and airborne systems since the 1990s has led to rapid technical progress. These new techniques have also been boosted by incredible progress in computing, which over the same period has allowed for both easier and cheaper calculations for radar studies. The kind of processing that used to require days of computation and an entire computer centre can now be done by anybody on a desktop PC. This has radically changed both data processing possibilities and the choice of algorithms. Our second reason was that we wanted to describe radarimaging techniques from a different viewpoint. Because there are so many different technical fields in which radar techniques are used, there is a correspondingly wide range of approaches. Specialists who have come to radar imaging with a background in electromagnetic signal processing techniques are likely to prefer a purely formal approach. Specialists in the use of radar images will see it as being derived from real aperture radar. Specialists in other kinds of image processing such as optical imaging or seismic tomography might prefer different approaches yet again. Our intention in this book is to maintain as geometrical an approach as possible to radar imaging. There are several different reasons for this. We believe that geometry enables us to use a more universal language than any of the specialised approaches that we have mentioned above. Although the geometrical approach may often seem naive, it nonetheless remains very precise. We thus see it as the simplest possible way of approaching the subject without sidestepping any of the difficulties and complexities inherent in radar imaging. Lastly, the geometrical results produced by radar make for the most spectacular applications. The most practical way of approaching this technique is therefore directly via those aspects which enable the widest range of radar imagery applications.
We have always noticed that our fellow radar specialists are particularly enthusiastic about their work. No one who has been involved with radar processing can ever quite leave it behind, despite any change in career. Of course, attempts to explain this must necessarily be partly arbitrary and personal. We believe however that this involvement depends on two factors. First of all, there is almost a rite of passage involved when confronting radar imaging, as it is necessary to make a leap of faith in order to understand the complexity of images which are purely computer-generated. While almost anybody can understand optical imagery, radar is harder to grasp; indeed, there is a significant barrier posed both by the abstract nature of the signal phase, as well as the elaborate reconstruction techniques employed. Those who then cross the barrier find themselves part of an elite club. Naturally, we do not claim that radar imaging is more complicated than other fields of observation techniques, at a specialist level, but the fact there is this barrier makes it appear far more complex. Furthermore, radar imaging has very fundamental links with the phenomena studied and the investigation methods employed. There is a striking similarity between the ambiguity in position and speed in radar imaging and Heisenberg’s principle of quantum mechanics, which also implies complex-based formalisms. There are also several analogies between radar polarimetry and quantum mechanics. It is precisely the fundamental concepts
We have divided this book into five parts. The first chapter (A Theoretical Emergency Kit for SAR Imagery) covers a few theoretical principles which by their very diversity place radar imaging techniques at the cross-roads between electromagnetism, signal processing and image processing. The propagation and polarisation of electromagnetic waves, the radiation of microwave antennas, the physics of radar measurements and the characteristics of the Fourier Transform are each dealt with in turn. The second chapter (SAR Processing: At the Heart of the SAR Technique) has nonetheless been written in such a way as to make it accessible to readers who are not familiar with remote sensing and who do not have previous knowledge of signal processing or radar physics. It describes radar processing in terms of geometry. The attentive reader should need no more than a basic scientific background. The third chapter (From SAR Design to Image Quality) deals essentially with radiometry aspects. Referring to these and to determination of the radar/target link budget it gives a detailed description of the decisive trade-off between the geometrical (resolution) and the radiometry (amplitude of the radar echo) features of a radar image. The fourth chapter (SAR Interferometry: Towards the Ultimate Ranging Accuracy) explains the principles and the main applications of radar interferometry which generally produces two types of information, most frequently combined in a single image: topographical information and information about ground movements. Finally, the fifth and final chapter (SAR Polarimetry: Towards the Ultimate Characterization of Targets) aims to explain the basics of polarimetry, which extends the possibilities of radar measurements by varying the polarisation.
The propagation and polarization of electromagnetic waves
A friendly warning to the reader: this discussion on the propagation and polarization of waves may bring back memories of school physics from another age as it does for the authors. The wave phenomena were profusely described at the time with images of ripples in water and skipping ropes. One day electromagnetic waves appeared on the black board. Newtonian physics suddenly faded into insignificance in the light of this new mysterious and infinite perspective. Indeed, for the first time, the teacher had referred to Maxwell’s equations with the fascination of a geographer for uncharted territories. According to him, they governed the behavior of this abstract universe, bedecked in symmetry.
Now, many years later, they remain as attractive as ever. Associated with constitutive equations and boundary conditions, they are the basis for this introduction. Their resolution is described for a standard case of propagation in a vacuum with no boundary conditions. An entirely deterministic plane wave structure is obtained, described by its polarization state. However the beautiful flow of this perfect wave is disrupted by the time or spatial fluctuations of the media it passes through. These perturbing encounters cause a depolarization effect and a dedicated formalism is needed to characterize it.