Neural system for tracking and classification of primary radar echo signals

Abstract

This work presents a system designed to acquire primary echo radar from aircraft and to operate a classification based upon spectral consideration of the retrieved echo. Fine spectral consideration is carried out exploiting the jet engine modulation (JEM) phenomenon due to the modulation on the primary echo introduced by engine rotating parts. Classification efforts are obtained by the use of a neural network previously trained, put on the system as a classification sub system. Target parameters are investigated analyzing effective signals obtained by two operating radars with an experimental system specifically designed for real time tracking of civilian aircraft using an acquisition sub-system, PC based, interfaced with the radar circuitry.