<p align="justify">The author makes a review of the SDR (Software Defined Radio) technology, including hardware schemes and application fields. A low performance device is presented and several tests are executed with it using free software. With the acquired experience, SDR employment opportunities are identified for low-cost solutions that can solve significant problems. In addition, a list of the most important frameworks related to the technology developed in the last years is offered, recommending the use of three of them.</p>

Resumen− La operación de los radares costeros y oceá-nicos se ve afectada porque los blancos se encuentran embebidos en un fondo de clutter marino. De acuerdo con el criterio de Neyman-Pearson, los detectores de radar siempre buscan garantizar un valor determinado de probabilidad de falsa alarma antes de mejorar otras variables del sistema. Utilizando la herramienta matemáti-ca MATLAB, los autores evaluaron el desempeño de los procesadores CA, OS, MSCA, AND, OR e IS-CFAR con respecto al mantenimiento de la probabilidad de falsa alarma concebida a priori en el diseño. Luego de someter los esquemas a diferentes perfiles de prueba con clutter distribuido Rayleigh, se concluyó que la mayoría de las alternativas presentan problemas ante determinadas situaciones que pueden aparecer con relativa frecuencia en ambientes reales. Consecuentemente, se ofrecen recomendaciones sobre cuál es el mejor esquema para emplear y garantizar una desviación reducida de la probabilidad de falsa alarma operacional con respecto a la de diseño cuando se enfrenta clutter heterogéneo.Palabras clave− Clutter de radar, CFAR, probabilidad de falsa alarma, desempeño de los detectores de radar.Abstract− The operation of coastal and off-shore radars is affected because the targets are surrounded by a background filled with sea clutter. According on the Neyman-Pearson criterion, radar detectors must always try to maintain a constant false alarm probability before trying to improve other system variables. Using the MATLAB mathematic software, the authors evaluated the performance of the CA, OS, MSCA, AND, OR and IS-CFAR processors concerning their ability to maintain the constant false alarm probability conceived in the design. After testing the schemes with different test profiles whose samples were Rayleigh distributed, it was concluded that most of the alternatives exhibit problems when facing certain situations that may appear in real environments. Consequently, recommendations on which solution is best to use are offered for guaranteeing a reduced deviation of the operational false alarm probability from the value conceived in the design when processing heterogeneous clutter.

A b s t r A c tThe performance of the CA-CFAR processor is affected by certain clutter variations. Although problems caused by sudden clutter changes have already been corrected in multiple CFAR proposals, the influence of slow statistical variations in the background signal is often ignored. To solve this problem, the authors estimated the optimal CA-CFAR threshold multiplier values necessary to adapt the processor to the clutter slow statistical changes. The application of the results guarantees that the operational false alarm probability of the processor will only exhibit a small deviation from the value conceived in the design. The clutter was simulated with a Pareto distribution with a known fluctuating shape parameter, according to recent papers that strongly suggest the use of this distribution. The current research completes an important step in the design of an adaptive detector that operates without a priori knowledge of the shape parameter. In addition, the authors provide mathematical expressions that allow the direct application of the results in the design of radar detectors. KeywordsConstant false alarm rate detectors, Pareto distribution, radar clutter, false alarm probability, adaptive selection of the threshold multiplier, CFAR processors. Estimación del Multiplicador Óptimo del Umbral CA-CFAR en Clutter Pareto de Parámetros Conocidos r e s u m e nEl desempeño del procesador CA-CFAR es afectado por ciertas variaciones del clutter. Mientras que los problemas causados por los cambios repentinos del clutter han sido corregidos por múltiples propuestas CFAR, se ignora frecuentemente la influencia de las variaciones estadísticas lentas de la señal de fondo. Para resolver este problema, los autores estimaron los valores óptimos del multiplicador del umbral CA-CFAR necesarios para adaptar el procesador a los cambios estadísticos lentos, garantizando por tanto que la probabilidad de falsa alarma del detector exhibirá solamente una ligera desviación con respecto al valor concebido en el diseño. El clutter fue simulado con una distribución Pareto con parámetro de forma conocido de antemano, de acuerdo a publicaciones recientes que sugieren fuertemente el uso de esta distribución. La investigación actual completa un paso importante en el diseño de detectores adaptativos que operan sin el conocimiento a priori del parámetro de forma. Adicionalmente, los autores proporcionan expresiones matemáticas que permiten la aplicación directa de los resultados en el diseño de detectores de radar. PAlAbrAs c l Av eDetectores de razón de falsas alarmas constante, distribución Pareto, clutter de radar, probabilidad de falsa alarma, selección adaptativa del umbral de detección, procesadores CFAR O desempenho do processador CA-CFAR está afectada por certas variações da desordem. Enquanto os problemas causados por mudanças bruscas de lixo foram corrigidos para múltiplas propostas CFAR, é muitas vezes ignorado a influ-ência de variações estatísticas lento do sinal de fundo. Para resolver esse problema, os autores estimaram...

No abstract

-The main problem faced by naval radars is the elimination of the clutter input which is a distortion signal appearing mixed with target reflections. Recently, the Pareto distribution has been related to sea clutter measurements suggesting that it may provide a better fit than other traditional distributions. The authors propose a new method for estimating the Pareto shape parameter based on artificial neural networks. The solution achieves a precise estimation of the parameter, having a low computational cost, and outperforming the classic method which uses Maximum Likelihood Estimates (MLE). The presented scheme contributes to the development of the NATE detector for Pareto clutter, which uses the knowledge of clutter statistics for improving the stability of the detection, among other applications.