Autonomous ESM Systems

Abstract

ABSTRACT Electronic Warfare (EW) Support Measures (ESM) systems are evolving rapidly toward unmanned status by using a constellation of artificial intelligence (AI) technologies. Among these are artificial neural networks (ANN), genetic algorithms, intelligent databases, fuzzy logic and data fusion techniques, as well as expert systems. The applications and implementation techniques, for which adaptive autonomous non cooperative target recognition is useful or necessary, form the focus of this discussion. Military and civilian applications abound for fixed site, shipborne, and airborne systems. These systems could control and manage other interactive systems normally requiring manual inputs. Applications such as air and marine traffic control and management regimes, collection of transport traffic statistics for economic analysis and intelligence, passive targeting, platform inventory (order of battle), and anti‐fratricide control are examples meeting this requirement.ESM systems use these AI technologies throughout the signal processing chain and beyond: from acquisition through automatic RF band scheduling and antenna pointing, as well as deinter‐leaving of dense signal environments and characterization of individual emitters, to rapid, high‐confidence emitter‐platform correlation with automatic optical/IR imaging, and cross‐indexing of multiple worldwide information databases to prevent platform aliasing. These systems are now feasible because the required components are now available. Low‐cost miniaturized RF system and subsystem components, massively parallel computing resources, and ANN integrated circuits permit the achievement of low‐cost, low‐power, small, modular systems that stand and act alone.