Adaptive radar pulse detection design based on difference of box filter

Abstract

Precise radar pulse detection is the premise of electronic support measures. A constant false alarm rate (CFAR) detection algorithm based on difference of box (DOB) filter is proposed to realise low-complexity and environment-adaptive radar pulse detection, and the principle behind the proposed algorithm is revealed. Specifically, the DOB filter is adopted to extract the rising edges and falling edges of radar pulses, and the signal variation law of probability density functions (PDFs) during the process of DOB filter are first theoretically analysed. Based on the derived PDFs, dynamic detection threshold and fixed threshold factor for the proposed CFAR algorithm based on DOB filter are deduced to detect the presence of radar pulses. Furthermore, the closed-form expression of the detection probability is derived. Simulations and on-board field programmable gate arrays implementation verify the superiority of the proposed CFAR algorithm based on DOB filter. Simulation results show that the proposed algorithm performs well in detecting various types of pulse signals, the detection probability ≥99% on condition that the signal-to-noise ratio ≥3 dB and false alarm probability = 10⁻⁸. Moreover, the proposed CFAR algorithm based on DOB filter can deal with both singular pulses and pulse on pulse signals.