Reverse Modeling and Autonomous Extrapolation of RF Threats

Abstract

This paper addresses the investigation of the basic components of reverse modeling and autonomous extrapolation of radio frequency (RF) threats in electronic warfare settings. To design and test our system, we first model RF threats using the radioactive parameters received. The enemy radar simulated with a transponder or emitter transmits electronic signals; next, the sensors of the system intercept those signals as radioactive parameters. We generate the attributes of RF threats during communication between the electronic emissions of RF threats and the receivers of our system in various electronic warfare scenarios. We then utilize the data acquired through our system to reversely model RF threats. Our system carries out the reverse extrapolation process for the purpose of identifying and classifying threats by using profiles compiled through a series of machine learning algorithms, i.e., naive Bayesian classifier, decision tree, and k-means clustering algorithms. This compilation technique, which is based upon the inductive threat model, could be used to analyze and predict what a real-time threat is. We summarize empirical results that demonstrate our system capabilities of reversely modeling and autonomously extrapolating RF threats in simulated electronic warfare settings.