Real-time simulation of electronic scanning antennas

This paper demonstrates an optimal marriage of two diverse technologies: electromagnetic antenna modeling and distributed (parallel) computing techniques. It presents a new and innovative methodology for modeling a number of antennas, with electronic scanning capability, in real-time, tactical airborne radar system simulations. The initial discussions compare execution time test results for antenna gain and phase-shift calculations, for several different antenna types. These include examples of electronically scanned acquisition radar and four-horn monopulse tracking antennas as well as mechanically-scanning antennas. This analysis defines the relative computational complexities for models of these types of antennas. Execution results are compared for three classes of algorithms: rigorous, empirical, and lookup tables. The algorithm classes were evaluated using a mainframe VAX and three grades of personal computer (PC) systems: (80)286, (80)386, and (80)486-based machines. The number of times the various antenna pattern algorithms could be interrogated in certain time periods are given, for each type of antenna and computer, using the types of antenna typical of both airborne and ground-based tracking and acquisition radars. These results define the relative, machine-dependent performance of the antenna modeling algorithm classes. Finally, the paper discusses means by which parallel, distributed computing techniques may be applied to speed up the execution of the algorithms.<>