This paper presents microwave tomographic reconstructions of the complex permittivity of lossy dielectric objects immersed in water from experimental multiview near-field data obtained with a 2.45-GHz planar active microwave camera. An iterative reconstruction algorithm based on the Levenberg-Marquardt method was used to solve the nonlinear matrix equation which results when applying a moment method to the electric field integral representation. The effects of uncertainties in experimental parameters such as the exterior medium complex permittivity, the imaging system geometry and the incident field at the object location are illustrated by means of reconstructions from synthetic data. It appears that the uncertainties in the incident field have the strongest impact on the reconstructions. A receiver calibration procedure has been implemented and some ways to access to the incident field at the object location have been assessed.

The paper is focused on the characterization of objects immersed in water using laboratory-controlled data obtained in the microwave frequency range. Experiments performed at the laboratory represent, at a reduced scale, the electromagnetic characterization of objects buried at a shallow depth in the sea, with the objects and the emitting and receiving antennas being immersed. Characterization is taken here as an inverse scattering problem whose data, whilst limited in aspect, consist of the values of the time-harmonic scattered electric fields measured at several discrete frequencies on a single line of a few receivers. The modelling of the wave-object interaction is performed through a domain integral representation of the fields in a two-dimensional transversemagnetic configuration. The inverse scattering problem is solved by means of two iterative algorithms tailored for homogeneous objects: the so-called level-set method and a binary specialized contrast source inversion method. Emphasis is put both on the experimental features and their modelling, and the results obtained for different types of objects are presented.

We are concerned, herein, with the conception and the design of printed UWB antennas and with the analysis of their performances through figures of merit. State of the art gives the opportunity to exploit some empirical considerations about the shape and the methodology of conception. A simple structure is simulated by means of CST Microwave Studio to show the possibility of fulfilling, at a defined level, UWB antenna requirements such as matching, fidelity and quasi-isotropy of the radiation pattern with a simple micro-strip design and a standard technology.

International audienceThis paper summarizes some tests with Low Frequency (LF, 125 kHz) RFID tags of two types: Card and Token. These tests were done in order to evaluate the feasibility of an identification/traceability of tags which size is constrained and supposed to be detected inside a delimited volume of 40x40x10 cm3. As the size of the antenna tag is supposed to be very small, we improve the detection range and volume of definition by designing different reader antennas. Reader antennas presented are of two types whether they are based on single (SL) or multiple loops (ML). Detection range was evaluated for planar antennas (3 SL and one ML). Volume of definition for the detection was estimated by designing two-level prototypes of ML antennas. Results are discussed about the optimization possibility of detection range and volume thanks to M

The specific properties of microwaves have been used for a while in many applications. More Reconstruction ~l g o r~t h m~The main difficulty in producing microwave images of quality was to compensate for complex diffraction mechanisms. Two classes of approaches have been, and are still, used. The fint one tries to take profit of available efficient reconstruction algorithms used with X-ray CT. X-rays propagate along linear paths, and using these algorithms in a diffraction situation requires to eliminate multipath contributions between the transmitter and the receiver. This has been achieved by using time spectrometry which adlows, thanks to a convenient time windowing, to isolate the shortest path corresponding to almosii straight line propagation [6,7]. To some extent, this fimt approach consists in a hardware attempt to compensate the effects of the scattering phenomena. In the second approach, on the contrary, the scattering mechanisms are formally taken into account in the reconstruction process, more or less rigorously. Indeed, two main approaches can be considered. The first one neglects multiple scattering in the target. This approach, imported froan ultrasound imaging techniques [8], is known as diffraction tomography. Such a lowdiffraction assumption allows to linearize the inverse scattering problem, in that sense that the scattered field data are linearly related ito the equivalent currents induced in the target by the interrogating beam. The reconstruction can be easily performed by using efficient Fast Fourier Transform algorithms, which can be viewed as a generalization of Fourier transform algorithms used with X-ray CT. Consequently, this approach is very efficient under the computational aspect and is particularly well suited for real time operations. However, the spectral processing neglects the evanescent part of the scattered field, and, as a result, the spatial resolution is limited to one half wavelength. Furthermore, this approach does not allow for quantitative imaging, because the reconstructed induced currents depend both on the local properties of tissues and on the total local field which varies within the target. The images obtained with diffraction tomography algorithms only reproduce the shapes of structures exhibiting low dielectric contrast, as for ultrasound echotomography in the case of soft tissues. The second approach, on the contrary, is able to provide quantitative images, even for high contrast targets. The images can be calibrated in terms of complex permittivity. With respect to previous diffraction tomography, the price to pay is very significant at the computation level. Indeed, an exact inversion of the scattering problem needs to iteratively solve a non-linear equation for the dielectric contrast. non-linear iterative techniqueAt each step of the iterative process, the measured scattered field is compared to the scattered field calculated from a numerical model. The permittivity contrast of the model is progressively adjusted by minimizing the error betw...

No abstract

No abstract

In this paper, the advantages of planar Microwave Tomography (MT) applied to early stage breast cancer detection are presented. In the proposed planar configuration, the breast is compressed between two dielectric plates in a configuration similar to that of X-ray mammography. This approach would allow the future implementation of a dual modality imaging system where the advantages of both techniques can be exploited. The research efforts made both at DRÉ/L2S (Supelec) and Poly-grames (École Polytechnique de Montréal), for the development of a planar MT system are described, as well as, the key features of the latter. A numerical validation is used to show how the breast compression can lead to an enhancement of the reconstructed images.