Abstract
In this paper we consider the problem of detecting range spread targets in the presence of Gaussian disturbance with unknown covariance matrix. To this end we resort to the modified generalized likelihood ratio test (MGLRT) and devise a robust detector, the orthogonal rejection MGLRT (OR-MGLRT), capable of deciding whether some observations contain a useful target or if they contain a signal belonging to the orthogonal complement of the useful subspace. Hence we investigate the possibility of exploiting the new algorithm as the second stage of a double threshold receiving structure, obtained cascading the MGLRT and the OR-MGLRT. At the analysis stage, we assess the performance of the new double threshold decision rule, both in the matched and mismatched signal cases, also in comparison with previously proposed receivers. The results show that the new test allows for a wide range of compromises between the detection and the rejection performance.

This publication has 16 references indexed in Scilit: