An extrapolation technique for antenna and RCS analysis involving large bodies

Summary form only given. The objectives of this paper are twofold. First we present a technique, based on the reciprocity principle, for computing the patterns of antennas located on large structures. Towards this end we utilizing the technique based on the fast multipole method (FMM), which enables us to handle problems that require tens of thousands of unknowns in the context of the method of moments. The FMM makes it possible for us to get to the lower edge of the high frequency regime, where the currents on the body begin to exhibit certain asymptotic characteristics when it is illuminated by a plane wave. We take advantage of this behavior of the currents to extrapolate them to higher frequencies for which the problem becomes too large to be handled, even by using the FMM. We illustrate the application of this approach by considering slot antennas mounted on cylindrical structures, and show how the pattern extrapolation for this type of problem is carried out. The second part of the paper also deals with the extrapolation approach, but for RCS computations. We show, once again, that the bistatic RCS of targets with multiple edges and corners can be tracked as a function of the frequency for a given combination of incident and observation angles to generate a representation for the scattered field that can be extrapolated to higher frequencies. Thus problems that are too large to handle by conventional numerical techniques now become tractable, once this representation covering the resonance region and the lower end of the high frequency range can be found. Towards this end we employ a combination of the generalized pencil of function (GPOF) method and the genetic algorithm (GA) to derive a finite series of complex exponentials, whose coefficients and exponents are both functions of frequency. This representation is initially matched over a certain frequency range, whose upper limit is the highest frequency for which the given problem is numerically tractable. Once a good match has been established, it is relatively straightforward to carry out an extrapolation of the frequency response to as high a frequency as is desired. The paper concludes with a discussion of the scope and limitations of the proposed extrapolation approach for solving high frequency scattering and radiation problems.