Search

The objective of this work is the analysis of the capacity of autonomous vehicles to position themselves in absolute terms in space (location) and the ability to observe their environment in order to move in it (environment detection). The project first performs a brief analysis of the resolutions that allow positioning systems external to the vehicle, basically from satellites to vehicles, and we contrast it with the positioning requirements. In the second and most important block of this project, the capabilities of radar systems to reconstruct their environment are analysed. For this we will analyse its most important parameters and we will obtain an approximation of its resolutions in the different frequency bands. In a third block we will obtain "almost realistic" images of objects that an autonomous vehicle can find in its navigation. We will study their longitudinal and transversal resolutions and we will delve into the recognition capacity of these objects. Finally we will draw conclusions and suggestions for potential future work

In this paper we present improved techniques of unwrapping and geocoding to obtain Digital Elevation Models (DEM). These techniques have been developed in the Teoria del Senyal i Comunicacions (TSC) department. Some results are presented too.

This paper shows the a student training RADAR computer to control it, simulated and introduced transformation of a non-coherent X-band RADAR in station. Using the RADAR P. P. I and a personal Rayleigh targets and clutter statistics are in base band. Raw video signals and severa! post-processing data can be represented.

An operational processor to obtain high resolution radar images of the Earth surface from an airborne platform is presented. The paper describes the problem, its geometry and the steps to obtain a reflectivity image from the echos received by an airborne SAR (Synthetic Aperture Radar) sensor, characterized by an unstable trajectory. The performance of the processor has been verified with real data from the DLR E-SAR system.

This paper presents an interferometric algorithm capable of working with SAR images obtained from measurements carried out with antennas that describe a circular trajectory. The antennas are positioned on a vertical mast at a low height. Therefore, the system is very interesting for surface scattering studies and buried objects detection due to the very high spatial resolution that C-SAR (Circular SAR) images present. A summary of the analytical formulation of the C-SAR processor and a complete description of the interferometric algorithm are presented along. Some experimental results obtained in anechoic chamber are also shown.

SAR Interferometry allows the obtention of digital elevation models (DEM) by combination of two synthetic aperture radar images acquired from slightly different viewpoints. ERS-1 SAR interferometry has already been validated over small areas. When processing wider areas, new adaptative filtering algorithms such as the short time Fourier transform are required. However, phase unwrapping is still the critical step. LMS an weighted LMS are two phase unwrapping algorithms which overcome some disadvantages of traditional residue based methods. Some results are shown obtained from ERS-1 images from Tarragona.

A polarimetric scatterometer is a short-range, low-power radar system capable of measuring the complete backscattering matrix of the surface it illuminates.

This document presents an algorithm for the reconstruction of 3D radar reflectivity images using spherical-wave ISAR techniques. The target is illuminated with an spherical wave and near-field backscatterig measurements are obtained using an spherical exploration geometry. The solution proposed by this method is described and the results of some simulations are shown for its validation.

A new method for motion compensation has been developed for radar imaging of moving targets (ISAR). Phase shifts (due to target motion) among adjacent pulses from a STEPPED FREQUENCY signal, are removed with a parabolic compensation method. The simulated ISAR images below show the improvement achieved after motion compensation.

A new method for motion compensation has been developed for radar imaging of moving targets (ISAR). Phase shifts (due to target motion) among adjacent pulses from a STEPPED FREQUENCY signal, are removed with a parabolic compensation method. The simulated ISAR images below show the improvement achieved after motion compensation.

The paper presents an evaluation of a high resolution technique intended to enhance the radar detectability of targets. The method is based on the coherent integration of the individually resolved target scatterers. The measured radar signatures of two scaled-down aircrafts are used to validate the proposed approach. Also the possibility of using this technique to identify the target has been addressed.

Este trabajo resuelve el problema de gran interés para la industria aeronáutica: el cálculo de la sección recta (RCS) de blancos radar complejos en tiempo real. Para ello basta una estación de trabajo con un acelerador grafico 3-d hardware en lugar de los super-ordenadores vectoriales que requieren los métodos clásicos. El método desarrollado totalmente original e innovador, consiste en procesar gráficamente una imagen del blanco presente en la pantalla de la estación grafica. Con ello se consigue que el tiempo de cpu sea independiente del tamaño y complejidad del blanco, con lo que su principal aplicación es la estimación rápida de la RCS de blancos radar complejos. Este método, al que hemos denominado "procesado grafico", implementa las aproximaciones asintóticas de alta frecuencia que permiten obtener las principales contribuciones a la RCS: óptica física, para la reflexión en superficies; condiciones de contorno de impedancia, para la reflexión en recubrimientos absorbentes radar (RAM), método de las corrientes equivalentes para la disposición en aristas y un método grafico de iluminación global (radiosity) para las reflexiones múltiples entre superficies. Los resultados de este trabajo hacen posible, por vez primera, el diseño interactivo de formas de baja RCS "stealth" con una estación de trabajo, por lo que se ha conseguido una herramienta software para la industria aeronáutica de gran eficiencia y potencialidad en sus aplicaciones.

The paper deals wilh the problem of radar delection of a target echo embedded in Weibull cluner and while gaussian noise. To derive a suitable processing scheme, the paper resons lo the general theory of radar delection which applies lo any type of probabiliry density function (PDF) and Aulocorrelation funclion (ACF) of the larget and clutter.

In this paper, the radar detection of rargers embedded in K_clurrer and whire gaussian noise is examined. The compound K disrribution model is a good modelfor observed echoes amplitude stadistics over a wide range of conditions. It allows to work with spatial and temporal correlared models. The paper resorts to the general theory of radar detection, and propose a processing scheme adapted to the compound K_disrriburion model.

This paper presents an introduction to the study and development of a new concept of Bistatic Radar Receiver, which eliminates any kind of direct microwave or optical fiber link between them. It is based on discret phase-lock techniques to acquire and maintain all the signal parameters when the transmitter illuminates directly the receiver system.

A polarimetric compact range RCS measurement system covering 8 to 40 GHZ is presented. The sensitivity is evaluated both theoretically and experimentally. Some RCS and ISAR results are included.

The design of a polarimetric Compact Range system covering 12-40 GHz is presented. Includes a preliminar error analysis, feeder design, and plane wave evaluation.

Processing radar cross section data is possible to image sorne properties of a target. The paper presents two radar imaging algorithms allowing the reconstruction of reflectivi ty distribution of the target or i ts shape. The methods have been evaluated by numerical simulations for different measurement conditions. Sorne encouraging results are presented.

This paper presents an exogenous model to generate K-distributed, coherent and correlated clutter. This model has been tested for exponencial and gaussian autocorrelation functions, showing good fits for autocorrelation factors less than 0.9.

En esta comunicación se introduce un nuevo esquema de detector no paramétrico para aplicación Radar y se discuten brevemente sus prestaciones. En el 1er apartado se justifica la aparición de los detectores no paramétricos como solución del problema de detección Radar en ambientes de estadística de ruido desconocida o variante con el tiempo. Seguidamente, se definen las hipótesis utilizando muestras de referencia y pulses múltiples. Se definen también dichas hipótesis en términos de rangos de las muestras, definiciones especialmente útiles por ser la base del detector no paramétrico de rango. En el apartado 3 se presentan la estructura localmente óptima y la estructura subóptima del detector de rango que se propone. Se termina esta comunicación comentando algunos resultados derivados del análisis del detector subóptimo de rango. Se presentan algunas gráficas, extraídas de (7), que ilustran los anteriores comentarios.

A Multiplier-Convolution-Multiplier (M.C.M.) processor behaves as a frequency-time converter. A computer simulation study is made to analize the performance of this device in a noisy environment. Own to its compressive feature, it is able to discriminate signals more efficiently than non-compressive receptors.

A canonical model is examinated for generating coherent Weibull vectors wi th a desired correlation matrix and specified parameters. Such vectors may represent samples of a correlated radar signal. Statistical and espectral estimates to value Weibull correlated vectors are also discussed.

This paper shows the influence of several loss factors on radar systems. It deals specially with those loss factors which are due to atmospheric effects, such as refraction, reflexion and the rain effects on radar system performance.

En el presente artículo se presenta en primer lugar la problemática en la detección de helicópteros y se propone un método para detectarlos basado en el efecto Doppler creado por el rotor principal del helicóptero. Para ello, se realiza un estudio analítico de los ecos ¡ procedentes del rotor principal. A continuación se ve cómo repercute 1 nefastamente el hecho de que el tiempo de observación del radar no sea infinito . Para paliar el problema se recurre a técnicas de estimación espectral . Finalmente, se presentan varias simulaciones vía ordenador que comprueban la validez del método propuesto .

Radar detection procedures involve the comparison of the received signal amplitud to a threshold. In order to obtain a constant false alarm rate (CFAR), an adaptative threshold must be applied reflecting the local clutter situation. This paper discusses the cell averaging and ordered statistic approaches for single receivers and presents a recursive formulation of an algorithm that permits for a system formed by N different CFAR detectors and a fixed level of false alarm in the data fusion center, to test the different intermediate fusion rules between the OR and AND rules, rules of rank K, with 1 = < K = < N, to obtain the rank K that maximizes the detection probability in the fusion center. The threshold values Ti of each CFAR detector, are calculated as a function of the probability of false alarm and decision rule (rank K) used in the fusion center. The algorithm also permits us to consider the receivers as a function of their quality or of the quality of the channel that connects them to the fusion center. The performances of the different procedures are shown in several figures.

The E.M.C. of a prototype for cylindrical microwave tornography is analized, the initial design has been improved and first results are presented.