Identification of Radar Emitter Type with Recurrent Neural Networks
Sabine Apfeld, Alexander Charlish, Gerd Ascheid
Abstract
In this paper, we present a method for the identification of different multifunction radar emitter types. It is based on Long Short-Term Memory recurrent neural networks and a previously published hierarchical modelling approach. This approach maps radar pulses to different levels of symbols which can be regarded as parts of a radar language. We evaluate our method with an example emitter that can make use of three different resource management techniques. The results show that it is possible to distinguish between radar types that mainly use the same emission parameters but differ in the resource management method.
Topics & Concepts
RadarComputer scienceCommon emitterIdentification (biology)Artificial neural networkRecurrent neural networkResource management (computing)Artificial intelligenceEngineeringTelecommunicationsElectronic engineeringDistributed computingBiologyBotanyWireless Signal Modulation ClassificationGeophysical Methods and ApplicationsElectrostatic Discharge in Electronics