Constrained Optimal Estimation: An Application to Target State Estimation in the Presence of Multipath Noise

Defending a ship under attack by an incoming cruise missile requires accurate estimates of the missile's dynamic state. However, obtaining accurate estimates from radar position measurements can be difficult if the missile is flying close to the earth's surface. This is because the radar measurements of low flying targets contain multipath noise. Multipath noise is a nonstationary highly correlated, large amplitude noise that can completely obscure the true measurements and make the target appear to be below the earth's surface. Faced with this type of measurement noise, unconstrained estimators can produce estimates with negative altitude and other physically meaningless characteristics. In this paper, we present an algorithm for a constrained optimal target state estimator that incorporates the physical limitations of the missile to obtain more accurate estimates of the missile's state in the presence of multipath noise.