Maximum likelihood geolocation using a ground moving target indicator (GMTI) report

The ground moving target indicator (GMTI) radar sensor plays an important role in surveillance and precision tracking of ground targets. In this paper, we develop the ML (maximum likelihood) estimator for the 3D target position using the range and azimuth measurements from a GMTI report, terrain data, and sensor position. A key contribution of this work is the inclusion of error models in the terrain data and sensor position in addition to the usual error models for the GMTI measurement. Our results indicate that neglecting the error in the terrain data can introduce a large error in the vertical component of the position estimate. Neglecting the error in the terrain data and sensor position introduces a negligible error in the horizontal position accuracy. The target position obtained from a deterministic approach is used as an initial estimate for the ML estimator. We present numerical results for the ML estimate of the target position and covariance using the terrain data, WGS84 EGM96 geoid undulation data, and simulated GMTI measurements.