Abstract :
Radar micro-Doppler (m-D) signatures are of great potential for identifying properties of unknown targets. All the techniques developed for extracting m-D features for the past decade rely primarily on the assumption that the time series of the signal contains at least one oscillation or more during the coherent integration time or imaging time. However, many applications in real-world scenarios involve short duration signals and often require the detection and the estimation of m-D characteristics. Short duration signals may contain only a fraction of an oscillation. In this study, the authors develop two techniques to estimate the m-D parameters from a fraction of the period. In these scenarios, the coherent integration will cover only 1/4 and 1/2 of the oscillation. The performance of the proposed methods are evaluated using both synthetic and experimental data.
