Tracking Process
Consolidation
CFAR
Association?
Prediction
Initiation
Termination
Formation
(Track States)
Rate Monitoring
Control
Figure 1: Tracking Process
Consolidation
CFAR
Association?
Prediction
Initiation
Termination
Formation
(Track States)
Rate Monitoring
Control
Figure 1: Tracking Process
Consolidation
Association?
Prediction
Formation
(Track States)
Rate Monitoring
Control
Figure 1: Tracking Process
Tracking Process
Depending on the antennas rotation rate and pulse repetition frequency of radar it will illuminate targets on a number of transmissions (see: dwell time and hits per scan in chapter “Basics”). In each range cell on each pulse period the signal processing decides whether a target is present, and if so produces a partial report, which defines the information available for that detection. The goal of the tracking process is the compression of radar data from multiple reports per target to a single data set for each target.
Detection Consolidation
Often, a target may produce many adjacent threshold crossings. The sets of threshold crossings (and their amplitudes) are then used to interpolation to determine a more accurate value of each measured observables. The methods may be:
- Moving Window (Interpolation)
- Centre of mass correlation (Weighted average)
- Comparison with database: Fit data to the expected angle beam shape (angle), pulse response (range), or Doppler filter shape (Doppler velocity)
Adaptive Area Constant False Alarm Rate (CFAR)
Because the distribution of radar cross-section of aircraft overlaps the cross-section distribution of birds (and other clutter targets), a significant number of these targets will pass mean level FARs, when they are implemented of an individual range-azimuth-Doppler cell. Use of Area CFAR techniques can mitigate this problem.
Track Initiation
A track is usually initiated after the detection of three on more scans of the radar to prevent excessive false track from being established. These detections are checked consistent motion along a reasonable trajectory and velocity profile of targets of interest.
Track Smoothing & Prediction
On the basis of a series of past detections, the tracker makes a smoothed (filtered) estimate of the target’s present position and velocity; and using this estimate, it predicts the location of the target on the next scan. The prediction may made by an α–β tracker; or a self learning Kalman-Filter.
Track Files and Track Updating
A master track file is kept of all track that have been initiated. The track file usually contains the following information:
- Position and amplitude, and Doppler velocity measurements of each detection and their time tag;
- Smoothed position and velocity information;
- Predicted position and velocity information at the time of the next track update;
- Track firmness (a measure of detection quality).
After a detection is associated with a track, the track file is updated.
Track Termination
If data from target is missing on a scan of radar, track may be “coasted”. If data from target missing for a number of scans, the track is terminated. The criterion for terminating a track varies for different types of radars.