Tracking a maneuvering target in the presence of false returns and ECM using a variable state dimension Kalman filter

In this paper, we show the performance of a variable state dimension Kalman filter for tracking a maneuvering target under the real-world conditions defined in the second benchmark problem. The second benchmark problem extends the first benchmark problem by including false alarms (FA) and electronic countermeasures (ECM). A modified version of the nearest neighbor PDA data association method of Fitzgerald (1986) is used to handle multiple measurement conditions. Adaptive waveform and dwell revisit time selection methods, and track filter coasting, are used to handle the uncertainties introduced by false alarms, missed detection, maneuvers, and ECM. Special features for integrating the adaptive methods into the variable state dimension filter are discussed. The results of this paper should provide a baseline to which other more sophisticated tracking and data association approaches can be compared.