In the context of the design and optimization of complex systems of communication or detection, the objective of the PIM team is to develop methods and models EM and also improve specific electromagnetic calculation and optimization tools for using the representation and understanding of physical phenomena resulting from the interaction of the waves with the environment. These activities are closely linked with the research themes of the two other teams in the MOM depertment (MF - Functional Materials and DIM - Devices & Multi-Physical interfaces).
The skills are divided specifically in modeling and electromagnetic calculation (techniques to improve the convergence properties, stability and accuracy of simulation tools), in asymptotic methods, digital, hybrid and their adaptation to the simulation of different problems in modeling and characterization of propagation phenomena of EM waves and finally oriented modeling 'system' requires a thorough knowledge of how microwave components and devices. These skills structure of the PIM team research in three areas described briefly below.
Modeling and electromagnetic simulation are essential to help in the design of complex devices and systems both used for observations / Applications in near field but also in far field.
Despite the rise of computers, the complexity of the problems to be solved is growing even faster and we face the limitations of the software currently available on the one hand and, on the other hand, faced with conflicting results for some complex structures that need to accurately simulate both the overall behavior of the structure and phenomena in great detail.
In the multi-scale scenarios, conventional computational tools are thwarted due to the emergence of poorly conditioned problems and often intractable. The development of well packaged formulations are stable, massively parallel and consistent multi-scale is one of the major challenges of the electromagnetic current calculation.
Research activities around electromagnetism and associated numerical calculation is divided into three main topics:
-The development of asymptotic approaches, empirical models, semi-empirical, allowing in particular to aim for real time applications, or hybrid methods to expand the scope of electromagnetic simulators for multiscale problems.
- Development of new numerical schemes and discretization to improve the numerical stability properties, packaging, convergence, accuracy or to reduce the complexity of calculation.
- multi-physical models that combine with other electromagnetic phenomena such as acoustics, thermodynamics or mechanics.
These guidelines are applied to numerical methods including solving equations is performed according to the formulation (integral, differential or variational) in the frequency domain (finite element method, method of moments) or time (TLM methods, FDTD and MRTD). There are cases limits the ends of the spectrum: the asymptotic methods such as GTD (ray method), physical optics when the wavelength is very small compared with the dimensions and the static or quasi-static for the opposite situation, equivalent current method to the consideration of diffraction phenomena ...
In this theme, our team focuses as priority axes, scientific developments on the modeling of the EM scattering by ocean surfaces observed at grazing angles and multi-static configuration taking into account the weather conditions, the influence of the presence of oil at sea, on the comparison of EM scattering models (rigorous methods, asymptotic methods, empirical and semi-empirical methods, ...) in bi-static configuration, on modeling of GPS or GALILEO links in marine or forestry environment and the target presence near a natural surface. This considering different types of sea surfaces with linear or nonlinear hydrodynamic phenomena. Different aspects of both numerical as well as applicative (with operational considerations) are considered.
Knowledge and fine modeling of the propagation channel are key steps in the design and optimization of systems for wireless communications, radar, remote sensing, radio navigation ...
Wireless applications require the use of a transmission channel whose characteristics have a strong impact on the propagation of electromagnetic waves (terrestrial or satellite links). It is therefore essential to have a knowledge of macroscopic propagation mechanisms to obtain an accurate modeling of spatial and temporal properties of the channel.
This modeling must take into account both deterministic components and both diffuse components are only approximations of reality. The aims are to obtain a better prediction of the channel needed for the radio interface design or allow the study of the capabilities offered by multi-sensor connections (MIMO).
The characterization of the propagation environment is also essential for modeling the knowledge that it must be independent of the antennas used. The characterization using conventional sounder can be expanded to benefit to a multi-sensor type of characterization, paving the way for localization techniques or new measurement techniques for monitoring the radio spectrum. Other applications also require the study and characterization of specific transmission media (radio frequency antennas for contexts of underwater observatories for example).
Furthermore, taking into account interaction of the waves with the environment (objects, atmosphere, nature of the surface over which the propagation is carried out, ...) is essential, for example, in the field of radar where the prediction of the propagation losses represents a key step in the detection process. This consideration may also be useful for locating electromagnetic sources …
Based on models developed under previous themes and experimental facilities available and / or data provided in the context of projects with partners, one objective of this axis is to develop systems for different applications and / or contribute by analyzing the different elements of the chain, to improve the performance of various systems (detection, monitoring, tracking, imaging, communication ...).
The developed hardware platforms are implemented with distinct objectives. They are used to validate the simulation platform (the implemented models, simulation techniques, algorithms, ...), to explore new technical possibilities or services or they serve to demonstrate the potential of new circuits or devices or new architectures.
To make a tool for development and effective optimization for designers a platform of simulation (or virtual platform) must reflect as closely as possible the real physical system. Simulation tools of type circuits / systems generally offer low computation time given the power of today's machines while the study of electromagnetic phenomena require simulation times can be significantly larger, especially when we consider of large structures.
Specific simulation methods are then implemented based on the problem addressed (eg rigorous methods or asymptotic) and simulation models compatible with the 'system' should be developed based on such macro-modeling is a method of behavioral description based on the development of mathematical models describing the relationships of the input-output systems. This model greatly reduces the simulation time compared to using physical models, electrical or electromagnetic.