Parameter estimation for space surveillance based on sparse reconstruction

Responding to the severe threat to the space activities from increasing low earth orbit (LEO) space debris, orbital object surveillance and cataloguing based on ground-based radar has arouse much attention from the world. However, the success of the space debris surveillance system strongly depends on the performance of parameter estimation of the radar system. This paper presents a novel method for space debris' motion parameter estimation based on the famous fence-type space surveillance radar system. Based on the observation that orbital debris are sparsely distributed in the observed fence, we formulate this problem as a sparse signal reconstruction problem with respect to an overcomplete dictionary, defined using three motion parameters. Simulation results show that the proposed method provides high estimation performance, which is robust to low Signal-to-Noise Ratio (SNR) and exhibits super-resolution properties.