Rao Test for Distributed Target Detection with Partial Observation

This paper considers the detection of a distributed target in the partial observation scenario, where only a set of receiving data with missing entries is available. The target is assumed to be obscured by complex Gaussian disturbance (clutter plus noise) with structured covariance matrix. A detection scheme based on the Rao test is proposed in this case. The unknown parameters in the detector are estimated resorting to the matrix completion (MC) technique, utilizing the specific structure of disturbance and the low rank property of clutter covariance matrix. Compared to traditional methods, this algorithm improves the estimation accuracy and furthermore achieves a better detection performance. Simulation results are presented to validate the effectiveness of the proposed method.