Detection of subspace distributed target in partial observation scenario with Rao test

Abstract

This paper deals with the problem of detecting a subspace distributed target obscured by disturbance consisting of clutter plus white noise. We focus on the partial observation scenario where some of the radar observations are missing, a phenomenon that usually caused by interference, spectrum sharing, compressed sampling, and so on. Detection strategies are established based on the Rao test. Specifically, we first derive the Rao test with the assumption that the disturbance covariance matrix under the null hypothesis is known. Then, the unknown covariance matrix in the test statistic is replaced with a suitable estimate to make the detector adaptive. At the estimation stage, two cases are considered, involving with and without disturbance only secondary data. The estimate of the disturbance covariance matrix is obtained by solving an optimization problem in the respective case that considers both the likelihood maximization and low-rank property of the clutter covariance matrix. Simulation results are presented to verify the effectiveness of the proposed method.