Joint Estimation of Location and Scatter in Complex Elliptically Symmetric Distributions

The joint estimation of the location vector and the shape matrix of a set of independent and identically Complex Elliptically Symmetric (CES) distributed observations is investigated from both the theoretical and computational viewpoints. This joint estimation problem is framed in the original context of semiparametric models allowing us to handle the (generally unknown) density generator as an infinite-dimensional nuisance parameter. In the first part of the paper, a computationally efficient and memory saving implementation of the robust and semiparmaetric efficient R-estimator for shape matrices is derived. Building upon this result, in the second part, a joint estimator, relying on the Tyler’s M-estimator of location and on the R-estimator of shape matrix, is proposed and its Mean Squared Error (MSE) performance compared with the Semiparametric Cramér-Rao Bound (SCRB).

     

Statistical analysis of achievable resolution limit in the near field source localization context

In this fast communication, we derive the statistical resolution limit (SRL), characterizing the minimal parameter separation, to resolve two closely spaced known near-field sources impinging on a linear array. Toward this goal, we conduct on the first-order Taylor expansion of the observation model a Generalized Likelihood Ratio Test (GLRT) based on a Constrained Maximum Likelihood Estimator (CMLE) of the SRL. More precisely, the minimum separation between two near-field sources, that is detectable for a given probability of false alarm and a given probability of detection, is derived herein. Finally, numerical simulations are done to quantify the impact of the array geometry of the signal sources power distribution and of the array aperture on the statistical resolution limit.     

Influence of device geometry on SOI single-hole transistor characteristics

SOI single-hole transistors have been fabricated by intentionally converting a quantum wire to an island connected to source and drain by two narrow constrictions. Two devices with different constriction lengths were investigated. It is found that slight differences in constriction lengths can lead to dramatic differences in device characteristics. For the device with short constrictions, periodic Coulomb oscillations are obtained and persist at temperatures in excess of 100 K. The physical origin of the tunnel barriers of the device has been analyzed experimentally and investigated theoretically based on the self-consistent numerical results of the Schrödinger and Poisson equations. The result indicates that lower ground-state energy for holes in the narrow constrictions serves as a potential barrier responsible for the periodic Coulomb oscillations. For the device with longer constrictions, aperiodic drain current oscillations are observed. The analysis of the experimental results shows that the quantum wire connecting source and drain is converted into at least three islands, probably due to the pattern-dependent oxidation effect. Consequently, the charging energy combined with the quantum confinement energy for the smallest island gives rise to aperiodic drain current oscillations.     

New experiments on the electrodeposition of iron in porous silicon

We report here the study about the electrodeposition of iron into porous silicon made in p-type (15–25 Ω cm) silicon wafers. Chronoamperometry measurements were performed to show that the iron nucleation does not start only at the bottom of the pores, which is confirmed by the high quality SEM images. The energy band of the heterostructure Si/PS is used to explain the mechanisms involved in the electrodeposition of iron and the porous silicon formation. This new structure (iron and porous silicon), once well controlled might have an influence on the new device developments.     

A Cramér Rao bounds based analysis of 3D antenna array geometries made from ULA branches

In the context of passive sources localization using antenna array, the estimation accuracy of elevation, and azimuth are related not only to the kind of estimator which is used, but also to the geometry of the considered antenna array. Although there are several available results on the linear array, and also for planar arrays, other geometries existing in the literature, such as 3D arrays, have been less studied. In this paper, we study the impact of the geometry of a family of 3D models of antenna array on the estimation performance of elevation, and azimuth. The Cramér-Rao Bound (CRB), which is widely spread in signal processing to characterize the estimation performance will be used here as a useful tool to find the optimal configuration. In particular, we give closed-form expressions of CRB for a 3D antenna array under both conditional, and unconditional observation models. Thanks to these explicit expressions, the impact of the third dimension to the estimation performance is analyzed. Particularly, we give criterions to design an isotropic 3D array depending on the considered observation model. Several 3D particular geometry antennas made from uniform linear array (ULA) are analyzed, and compared with 2D antenna arrays. The isotropy condition of such arrays is analyzed. The presented framework can be used for further studies of other types of arrays.     

Weiss-Weinstein bound for MIMO radar with colocated linear arrays for SNR threshold prediction

Several works have suggested that a multi-input multi-output (MIMO) radar system offers improvement in terms of performance in comparison with classical phased-array radar. However, under the widely spread assumption of a uniform a priori distribution for one parameter of interest, there is no result concerning lower bounds on the mean-square error in the case of a Gaussian observation model with parameterized mean. This Fast Communication fills this lack by using the Weiss-Weinstein bound (WWB) which can be calculated under this difficult scenario. As we will show, the proposed bound for MIMO Radar with colocated linear arrays has no closed-form expression. To solve this problem, we propose a closed-form approximation that, as we will show by simulations, is close to the actual bound. This approximated bound is then analyzed for a design purpose in terms of array geometry. Simulations confirm the good ability of the proposed bound to predict the mean square error (MSE) of the maximum a posteriori (MAP) in all ranges of SNR. Particularly, the tightness of the bound to predict the SNR threshold effect is shown.     

LDMOS in SOI technology with very-thin silicon film

In this paper, we extensively investigate, by two-dimensional simulations, the output characteristics accuracy and breakdown voltage performance for very-thin film (80 nm) SOI lateral double-diffused MOS (LDMOS) transistor as a function of the drift doping, drift length and field plate length. Trade-offs are discussed to optimize the off-state breakdown voltage versus the occurrence of kink effect and quasi-saturation in on-state. The conclusions are supported by experimental results.     

A CramÉr-Rao Bound Characterization of the EM-Algorithm Mean Speed of Convergence

This paper deals with the mean speed of convergence of the expectation-maximization (EM) algorithm. We show that the asymptotic behavior (in terms of the number of observations) of the EM algorithm can be characterized as a function of the Cramer-Rao bounds (CRBs) associated to the so-called incomplete and complete data sets defined within the EM-algorithm framework. We particularize our result to the case of a complete data set defined as the concatenation of the observation vector and a vector of nuisance parameters, independent of the parameter of interest. In this particular case, we show that the CRB associated to the complete data set is nothing but the well-known modified CRB. Finally, we show by simulation that the proposed expression enables to properly characterize the EM-algorithm mean speed of convergence from the CRB behavior when the size of the observation set is large enough.     

Effect of the crystalline constitution of TiO2 substrates on the growth of ultrathin Mo layer

Metallic molybdenum was deposited by magnetron sputtering on amorphous and (110) rutile TiO₂ substrates. An interfacial reaction between the deposited Mo and the TiO₂ substrates generating Ti^3 +, Ti^2 + oxidation states is evidenced by X-ray photoelectron spectroscopy. Our XPS data suggest, as compared to the (110) rutile substrate, a higher reactivity of the amorphous TiO₂ leading to a stronger Mo oxidation. In both cases, this reaction, leads to the formation of MoO_x nanostructures at the interfaces. The growth mechanism of the Mo deposit as a function of the crystalline constitution of the TiO₂ substrate was analyzed by processing the XPS data using the Quases ® software. The data reveal a layer-by-layer growth of the Mo deposit on the (110) rutile substrate and a Stranski–Krastanov growth on the amorphous one. We explain these different growth modes based on the TiO₂ surface reactivity and electronic structure using the Cabrera–Mott theory. This explanation is supported by Time-of-Flight Secondary Ion Mass spectrometry profiling.