A neural network approach towards multiradar track fusion

The Jindalee operational over-the-horizon radar network has three radar stations which overlap in their coverage. Consequently, targets may be detected by more than one radar. Multiple tracks from one target present a large operator workload, especially at where regions overlap. This paper introduces a method to automate the track association process. Multilayer perceptron and statistical decision systems are used to classify tracks into associated and nonassociated pairs.