KB-GLRT: exploiting knowledge of the clutter ridge in airborne radar

Knowledge-based radar detection for space-time adaptive processing applications is addressed. At the design stage, the characteristic structure of the clutter ridge was taken account of and two decision rules were devised according to the generalised likelihood ratio test (GLRT) and the two-step GLRT criteria. The authors first deal with the case of a clutter ridge with integer slope and then discuss the more general framework of a non-integer slope. For the latter case only approximate GLRT detectors are provided due to the analytical difficulties connected with the exact solution of the problem. Analysis shows that the new knowledge-based systems can outperform some previously proposed adaptive schemes and can achieve a performance level very close to the optimum detector, which assumes the perfect knowledge of the clutter covariance matrix.