An algorithm for estimating parameters of a group of closely spaced random processes: Synthesis, analysis, and modeling (Q1286548)

The paper is devoted to the estimation of the track parameters of closely spaced targets moving in a space where the coordinates are measured directly. The observation conditions are assumed to be such that probabilities of disappearance of a mark or appearance of a false mark are negligible. Under these conditions, the main problem of estimating track parameters consists in comparing the marks given to the observed target in the current frame. A Bayes method of solution is chosen. It is demonstrated that this algorithm is branching and its realization in practice is very difficult. In addition, to an optimal algorithm, a quasi-optimal one is considered. Some considerations on the practical realization of modified Bayes algorithms and on the results of numerical modeling are presented. The problem is solved in discrete time.