From the waveform through the resolution cell to the tracker

Waveform selection problems and some sidelobe-reduction techniques are investigated from a system engineering point of view via the hybrid conditional averaging (HYCA) method. A probabilistic data association filter (PDAF) is assumed in the HYCA method. The resolution cells are defined as tesselating parallelograms with constant volume, which are oriented and shaped to match the ambiguity function. The detection model is improved by considering the effect of detections in neighboring cells caused by ambiguity function sidelobes. Both Swerling 1 and Swerling 3 target models are investigated. Results indicate that up-sweep linear FM (LFM), and 50%-50% upsweep LFM with low sweep rate/CF combination, are both competitive choices. As far as the two kinds of sidelobe-reduction techniques are concerned, Hamming windowing is found preferable to Gaussian shading. The effect of the detection threshold is also studied. For the constant volume of the resolution cell, we take that volume as natural for the CF waveform. From the coded pulse results, it appears that this volume is an important parameter to be varied in future investigations