An approach to the study of automatic target tracking

Several problems encountered in the design of automatic processing systems for data obtained from search radars give rise to a requirement for the prediction of the present or future position of a target on the basis of past position reports. The most obvious of such problems is that of scan-to-scan correlation of target reports. Here the prediction of the present position of the target determines the center of the “gate” in which the target is sought; and the accuracy of this prediction determines the size of the gate. It is important that the prediction be accurate and the gate size reasonably small in order to avoid confusion between neighboring targets. Any scan-to-scan correlation system will break down when the density of targets becomes too great. The accuracy of prediction determines the density at which trouble may be encountered. While it is not expected that target densities will ordinarily attain unmanageable proportions, this situation could easily arise due to enemy action in launching decoys and chaff.