Low Complexity Detection Scheme for Fast Target in Randomized Stepped Frequency

Abstract

In randomized stepped frequency radar, two-dimensional matched filtering is performed to detect the target and estimate the range and radial velocity of target. However, if the region of target velocity to be considered is very wide, thus, target maneuvers very quickly, matched filtering should be performed to the corresponding wide velocity region. Then, grating lobe occurs periodically in the velocity axis, thereby causing a problem of finding a mainlobe in the grating lobe. In this study, we propose a two-stage low complexity detection algorithm for high-speed target. In the first stage, the coarse detection which determines the velocity region including mainlobe is performed. Then, fine detection is processed by performing 2D matched filtering in the reduced range-velocity region in the second stage. By dividing the target detection into two steps, computational complexity can be drastically reduced. The performance of the proposed technique is verified through simulation results.