Application of pso to electromagnetic and radar-related problems in non cooperative target identification

Particle Swarm Optimization (PSO) is a technique widely used in many applications areas and is being studied extensively by many researchers. Regarding electromagnetism, PSO [1] has shown to be one of the algorithms with the best behaviour when facing several specific problems. In this contribution, the application of PSO to two real problems is presented as a part of our work in computational electromagnetics applied to Non Cooperative Target Identification (NCTI). These are the estimation of the Direction of Arrival (DOA) of various incoming signals and the estimation of the electromagnetic constitutive parameters of dielectric materials. DOA estimation [2] is a well known problem involved in different applications. For instance, it is used in ESM (Electronic Support Measurement) systems to determine the angular position of a target from its emissions. For the next generation of Passive Radars using Phased Array Antennas, the determination of the target position should be done by these techniques too. In this sense, if we want to be able to perform the identification by means of NCTI techniques it is very important to understand and master these techniques. Due to characteristics of these new radars, new efforts are necessary for estimating the DOA with fewer snapshots. In that sense, PSO provides a simple but fast and accurate solution to the DOA problem using a single snapshot. The application of PSO to DOA will be presented in full detail, including also a performance study of the algorithm. Apart from the DOA techniques, the estimation of electromagnetic properties of materials is a key part of the process to generate a trustable model to be simulated by electromagnetic software tools in order to obtain its radar signature for identification purposes. This estimation can be accomplished through free space reflection measurement of the material in an anechoic chamber. Formerly, the classical methods tried to estimate the complex permittivity (ε*) from the measurements minimizing an error function based on the transmission line theory. Nowadays, soft-computing techniques like PSO are beginning to be used, with accurate results.