COMPUTER ARCHITECTURE AND VERY HIGH-LEVEL LANGUAGE DESIGN FOR BATTLE MANAGEMENT

CURRENT COMPUTATIONAL APPROACHES ARE INADEQUATE TO PERFORM REAL TIME PATTERN RECOGNITION AND ADAPTIVE RESOURCE ALLOCATION IN ORDER TO SUITABLY RESPOND TO ENEMY THREATS WHEN COMPARED TO THE DYNAMICALLY CHANGING MILITARY ENVIRONMENTS. DESPITE DECADES OF EFFORT BASED UPON STATISTICAL SIGNAL PROCESSING AND PATTERN RECOGNITION METHODOLOGY, THEREAL TIME RECOGNITION OF COMPLEX TARGETS REMAINS A CHALLENGING PROBLEM. IN THE AREA OF BATTLEFIELD SITUATION ASSESSMENT, THERE DOES NOT APPEAR TO BE A SUITABLE APPROACH TO RESPOND TO THE COMPLEX DYNAMICALLY EVOLVING SCENARIO THAT SUCH ASSESSMENT AND ALLOCATION PROBLEMS REQUIRE. THE FEASIBILITY IS BEING EXAMINED OF DESIGNING A CONTINUOUS REAL TIME PATTERN RECOGNITION AND THREAT RESPONSE CAPABILITY IN DYNAMICALLY CHANGING MILITARY ENVIRONMENTS USING THE INNOVATIVE APPROACH OF A COMPUTATIONAL PERCEPTUAL PROCESS BASED ON A COGNITIVE NET. THIS STUDY PROVIDES AN OPPORTUNITY TO ATTACK VERY SIGNIFICANT DEFENSE PROBLEMS THAT HAVE LONG AWAITED PRACTICAL SOLUTIONS BY INTEGRATING IDEAS AND METHODOLOGIES DERIVED FROM COGNITIVE SCIENCE, NEURAL NET MODELING, PATTERN RECOGNITION, ARTIFICALINTELLIGENCE, AND NOVEL COMPUTATIONAL ARCHITECTURES USING CURRENTLY DEVELOPING TECHNIQUES SUCH AS VERY LARGE SCALE INTEGRATION.