Mariano Raboso

Acoustic Biometric System Based on Preprocessing Techniques and Linear Support Vector Machines.

Drawing on the results of an acoustic biometric system based on a MSE classifier, a new biometric system has been implemented. This new system preprocesses acoustic images, extracts several parameters and finally classifies them, based on Support Vector Machine (SVM). The preprocessing techniques used are spatial filtering, segmentation—based on a Gaussian Mixture Model (GMM) to separate the person from the background, masking—to reduce the dimensions of images—and binarization—to reduce the size of each image. An analysis of classification error and a study of the sensitivity of the error versus the computational burden of each implemented algorithm are presented. This allows the selection of the most relevant algorithms, according to the benefits required by the system. A significant improvement of the biometric system has been achieved by reducing the classification error, the computational burden and the storage requirements.

Analysis and design of multifunction radar task schedulers based on queue

Inside the set of systems that constitute the avionics, the radar on board is one of the principal systems, both for the commercial and the military aircraft. For a few years ago, the technologies of electronic exploration arrays are used in the radar design, and specially, in the combat fighters, where the detection and tracking of multiple targets is a fundamental requirement. In this type of environments, it is required to use multifunction radar, MFAR (Multi-Function Array Radar), which joins inside the same system, and simultaneously, so much the classic functions of tracking and surveillance, as all the functions related to the communication, countermeasures, calibration, etc. Thus, the functions are implemented according to specific tasks. The principal ones are: surveillance, tracking, confirmation of false alarm, backscanning, reacquisition and communications plane-missile. Therefore, it is required to work with, specialized subsystems inside the radar. They are called task schedulers. The task scheduler is a key element of the radar, since it does the planning and distribution of energy and time resources to be shared and used by all tasks. This paper analyzes the features of the task schedulers based on tasks queues. Radar time is divided in time intervals that are called scheduling intervals. They allow realizing the task scheduling in a flexible and automatic way, planning individually each interval. Therefore, the task scheduler constitutes, for every scheduling interval, the corresponding queue or queues with the tasks planned to execute in that interval. Then, the tasks that are going to execute are selected from those tasks queues. Therefore, the scheduler includes and applies two scheduling policies: the policy for the constitution of the tasks queues, and the policy of scheduling, which is applied for planning every scheduling interval. Several schedulers have been designed and studied, and it has been made a comparative analysis of different performed schedulers. The tests and experiments have been done by means of system software simulation. Finally a suitable set of radar characteristics has been selected to evaluate the behavior of the task scheduler working.

Analysis of Directive Sensor Influence on Array Beampatterns

Over the past few years, a large number of pattern synthesis techniques of antenna arrays have been studied and developed. Such techniques may be classified into two categories: techniques that optimize the excitation (amplitude and phase) of each element in a uniform array (Van Veen & Buckley, 1988), and techniques that adjust the positions of the elements with uniform excitation, resulting in a non-uniform geometry (“Unz, 1960”, ”Harrington, 1961”, ”Skolnik et al., 1964”, ”Haupt, 1994”). Despite of this classification, both categories are not exclusive; so, it is possible to develop techniques that optimize both the excitations and the positions of the elements (”Akdagli & Guney, 2003”, “Kurup et al., 2003”, “Kumar & Branner, 2005”). It has been also observed that many of these techniques make a beampattern synthesis only in the case of an array pointing to the broadside. Only a few techniques are designed taking into account other angles further than the broadside (“Bae et al., 2005”, “Bray et al., 2002”, “Feng & Chen, 2005”), which is the basis of beamforming. The reason is that, these techniques work on the assumption that the array is formed by omnidirectional sensors. In this case, working with the array pattern in the u domain (u=sen(θ)), a variation of the pointing angle only implies that a shift in the beampattern, without variation of the characteristics of neither the main lobe, nor the sidelobes (Mailloux, 2005). Thus, representing the pattern in the u-u

Experimental Calibration for Electronic Beamforming with Sensor Arrays

A sensor array system with N channels is assumed to have the same characteristics for each channel, which are composed of a sensor and an amplification system. In beamforming applications, the gain and phase of each channel are key elements in the synthesis of the beampattern (van Veen & Buckley, 1988). On the other hand, the position of the sensors in the array and the orientation of the axis/plane where the array is placed are also important for an accurate calculation of the weight vector.

Virtualizing Grid Computing Infrastructures into the Cloud

This paper shows how virtualization techniques can be introduced into the grid computing infrastructure to provide a transparent and homogeneous scientific computing environment. Today’s trends in grid computing propose a shared model where different organizations make use of a heterogeneous grid, frequently a cluster of clusters (CoC) of computing and network resources. This paper shows how a grid computing model can be virtualized, obtaining a simple and homogeneous interface that can be offered to the clients. The proposed systemis implemented on a system named virtual grid. Both cloud computing infrastructure and grid computing technology used, are freely available to all users.

A Scientific Computing Environment for Accessing Grid Computing Systems Using Cloud Services

This paper shows how virtualization techniques can be introduced into the grid computing infrastructure to provide a transparent and homogeneous scientific computing environment. Today’s trends in grid computing propose a shared model where different organizations make use of a heterogeneous grid, frequently a cluster of clusters (CoC) of computing and network resources. This paper shows how a grid computing model can be virtualized, obtaining a simple and homogeneous interface that can be offered to the clients. The proposed system called virtual grid, uses virtualization support and is developed from integration of standard grid and cloud computing technologies.Furthermore, a Scientific Computing Environment (SCE) has been developed to provide uniform access to the virtual grid.

MATLAB COM Integration for Engineering Applications

The most powerful idea around component-based software, is that components can be implemented by a programmer and reused by others without having knowledge of the source code. Components are binary packages that can be deployed and further integrated with others written on different programming languages. As component selection and integration is usually an easy and well-known process, components are also called COTS (Commercial Off-The-Shelf).

Parameter Analysis of a Genetic Algorithm to Design Linear Array Geometries

This article summarizes several analyses on employing an iterative learning method of the Computational Artificial Intelligence field, a Genetic Algorithm, focused on designing linear arrays. The objective of these analyses is the effectiveness improvement of these evolutive algorithms in this particular problem. The influence of giving certain values to each of the specific parameters of a Genetic Algorithm is characterized. Obtaining the optimal final solution depends on these parameter values. Thanks to this analysis, the Genetic Algorithm is optimized and also the best linear array geometry, based on certain established quality criteria, is found.

Sidelobe Evaluation of Cardioid-Patterned Sensor Array

Due to the widespread use of sensor arrays, a great variety of design techniques have been developed, adjusting the excitation of the sensors and/or their positions. It has been observed that most of these techniques work under the assumption of invariance in the beampattern, in the sen (thetas) domain, of arrays formed by omnidirectional sensors. Therefore these design techniques are only defined for arrays pointing to broadside. As directional sensor arrays are highly used, invariance hypothesis is no longer valid. This paper studies the influence of using directional sensors in the array beampattern, to show if the effects caused on the pattern should be taken into account when designing such arrays.