AN IMPROVED TWO-DIMENSIONAL CFAR DETECTOR FOR MOVING TARGET DETECTION IN COMPLEX GROUND CLUTTER SCENARIOS

The CFAR detectors have been used widely in radar detection. Multi-dimensional CFAR detectors outperform one-dimensional CFAR detectors in many scenarios. CFAR detectors achieve optimal performance when the cells in the scenario are independent and identically distributed (I.I.D.). In scenarios with severe ground clutter, pulse-Doppler (PD) radar usually employs a high-pass finite impulse response (FIR) filter and coherent integration to detect moving targets. The filter has a limited passband width, breaks the I.I.D. in the Doppler dimension, and restricts the usage of multi-dimensional CFAR. This paper proposes a two-dimensional CFAR detector that is suitable for PD radar detecting moving targets under heavy ground clutter scenarios. It incorporates the magnitude response of the filter to the reference cells selection of the CFAR detector. These cells are more likely to be I.I.D. and can be used to estimate the interference level across Doppler bins. This detector does not require massive calculations. Based on a set of measured data with spatiotemporally varied ground clutter, this detector demonstrates more stable detection performance and fewer false alarms compared to traditional one-dimensional CFAR detectors.