Time-frequery image processing based jet engine modulation extraction method towards target recognition

Jet Engine Modulation (JEM) characteristics are widely used in airborne radar to detect and classify helicopter, propeller-driven aircraft and jet aircraft. Airborne clutter and low signal-to-noise ratio (SNR), however, veils the JEM phenomena. Factors such as Observation Angle, Blade Pitch, Intermodulation of Rotors and Radar Cross Section (RCS) scintillation, can affect the JEM characteristics of radar echoes significantly. Thus a robust and optimized JEM extraction method is in sore need. In this paper, we propose a fusion detection method, using super resolution to obtain JEM time-frequency images. By means of pre-processing, time-frequency analyzing and fusion detecting, JEM characteristics areextracted from complex background. Further measurements and data analysis verify its effectiveness. We anticipate that this method will form the basis for the development of JEM engineering and indicate possible lines for airborne radar’s target classificationin future.