A Model for Aerial Surveillance of Moving Objects When Errors of Observation Are Multi-Variate Normal.
Ingram Olkin, Saunders,Sam C · Defense Technical Information Center (DTIC) · 1978
This paper presents a general theorem on the invariant behavior of a certain function of a matrix. It then shows the importance of this result principally by using it to derive properties of certain maximum likelihood estimates which arise when considering problems such as the location of a moving object being surveyed from a moving observatory when all data on location are subject to stochastic error. This problem is important in tracking objects either from an observatory satellite or from a transport plane bearing ground seeking radar. Some applications to this situation are made. (Author)