Classifying Pulse Compression Radar Waveforms Using Time-Frequency Distributions
Jarmo Lundén, Visa Koivunen, Liisa Terho · 2005
In this paper the classification of pulse compression radar waveforms using features extracted from the Choi-Williams time-frequency distribution is studied. In addition, a feature based on the symmetry properties of polyphase waveforms is introduced. The pulse compression waveforms examined are the Frank, P1, P2, P3, and P4 codes. The discrimination capability of the features is evaluated using an ensemble averaging early-stop committee of 10 multilayer perceptrons. The classifier achieves over 96 % overall correct classification rate in signal-to-noise ratio of 3 dB on data similar to the training data.