Geolocating a remote emitter
A method for improving geolocation accuracy in a passive radar warning receiver, using synchronized data curve-fit and interpolation to asynchronous and noisy receiver and navigation measurements over observation periods that are extended to reduce inaccuracies caused by noise. The present disclosure yields synchronized data samples at intervals short enough that constant-rate equations are valid, even though the actual motions over the observation interval may be more complex and have higher-order dynamics. It reduces noise, synchronizes data samples, and is readily adapted to motions with variable acceleration. The method generates rate samples short enough to satisfy constant rate assumptions, yet fit data over intervals long enough to enhances measurement accuracy by reducing measurement noise.
1. A method of geolocating a remote emitter, the method comprising:
detecting, at two antennas positioned at respective locations on a platform, signals from the remote emitter along with their respective times of arrival over a series of observation periods;
calculating, by a radar warning receiver (RWR) coupled to the two antennas, an estimate of the remote emitter geolocation based on signal amplitude measurements of the detected signals;
selecting, by the RWR, a time equation f(t) representing TDOA variation of the detected signals corresponding to a differential motion of the two antennas over each of the observation periods, the differential motion comprising: a substantially constant acceleration of the antennas, a substantially constant velocity of the antennas, a sinusoidal motion of the antennas of a certain frequency, or flexure or vibration characteristics of the platform identified in TDOA data corresponding to the signals;
determining, by the RWR, equation coefficients for the selected time equation f(t) such that the time equation f(t) best fits the measured differential time of arrival of the detected remote emitter signal;
determining, by the RWR, equation coefficients for a time-derivative equation f′(t) of the time equation f(t) with the determined coefficients;
determining, by the RWR, calculation sampling times over an interval at which data samples will be generated;
determining, by the RWR, measured remote emitter signal frequency difference of arrival (FDOA) values from the time-derivative equation f(t), and/or measured remote emitter signal time difference of arrival (TDOA) values from the time equation f(t), at the calculation sampling times;
measuring, by inertial and location sensors provided on the platform, platform position and motion data, comprising motion, attitude, and position data of the two antennas at corresponding motion sampling times over a time frame generally preceding and including each observation period of the series of observation periods;
determining, by the RWR, FDOA and/or TDOA values at the motion sampling times, based upon the estimated remote emitter geolocation and the measured platform position and motion data;
calculating, by the RWR, by time interpolation of the determined FDOA and/or TDOA values at the motion sampling times, predicted FDOA and/or TDOA values at the calculation sampling times for each observation period;
comparing, by the RWR, the measured and predicted FDOA and/or TDOA values corresponding to the calculation sampling times, using one of: a Kalman process, or a least mean square process and generating an error value; and
modifying, by the RWR, the estimated remote emitter geolocation as a function of the generated error value.
2. The method of claim 1 , wherein motion sampling times are times used for measuring FDOA and/or TDOA values from the step of detecting.
3. The method of claim 1 , further comprising:
for each of the measured and predicted FDOA and/or TDOA values, the RWR calculating estimated noise as an RMS sum over the subject observation period between sampled data and a curve fit of the sampled data; and
using the RMS sum for noise-weighting of the generated error value, thereby reducing solution errors due to measurement noise.
4. The method of claim 1 , wherein the time function f(t) is a polynomial function.
5. The method of claim 1 , wherein the platform is one of: an aircraft, a ship, a missile, and a spacecraft.
6. The method of claim 1 , wherein the two antennas are adapted to detect electromagnetic radiation and the remote emitter is a radar source.
7. The method of claim 1 , wherein comparing the measured and predicted FDOA and/or TDOA values further comprises comparing current and previous measured and predicted FDOA and/or TDOA values over each of one or more observation intervals.