Multitarget tracking in clutter: two algorithms for data association

摘要

In this paper two algorithms for data association in the context of multiple target tracking on non-manoeuvering aircrafts using back-propagation neural network (BPNN) and learning vector quantization neural network (LVQNN) are presented. The performances of the algorithms are compared with those of the standard method for data association based to the nearest-neighbour rule by Monte Carlo experiment and by using real radar data records.