Navigation: Synopsis, Estimation Theory, Radar Trackers, Data Fusion, Top
Link to FAA Tech center course home page: Click Here (opens in a new window)
Formal Synopsis: Click here (PDF)
This is an adaptation of a course that was taught by Dr.
Polikar and I at the
Navigation: Synopsis, Estimation Theory, Radar Trackers, Data Fusion, Top
The first topic, Estimation Theory, begins with application of probability and statistics and a refresher on elementary linear algebra to form a basis for understanding how multiple measurements can be used to estimate multiple parameters. An example is how Doppler radar and optical measurements of target azimuth can be used together to estimate the range and velocity of something moving at constant speed, something that neither measurement can do separately. This segment will end with an overview of Kalman filters and applications of Kalman filters.
Navigation: Synopsis, Estimation Theory, Radar Trackers, Data Fusion, Top
The second module, Radar Trackers, presents a variety of trackers, most based on a Kalman filter. The radar tracker is presented as a system or software function that serves to update a database of radar contact positions and other data with new information from radar returns. Radar platforms such as ground-based air surveillance and air defense, air-base radars, and space-based radars are discussed. Ground-based radars are mainly used as a context for most material. This function has an architecture that is determined to serve its functions best with given input data and requirements of the database, and often consists of several estimation algorithms of various types. Most of these can be characterized as Kalman filters but others are estimators of fixed parameters using blocks of data (batch estimators). Radar trackers can also accept data from outside sources, which opens us up to data fusion concepts.
Navigation: Synopsis, Estimation Theory, Radar Trackers, Data Fusion, Top
The third module, Data Fusion, shows how data from multiple sources can be used to provide far more information than can be provided by any single data center. Sensor fusion underlies most new large system concepts. The DoD has a well-defined infrastructure, and this is presented as background. As a simple example of data fusion, international maritime authorities, including the U.S. Coast Guard, require something called Automatic Identification System (AIS). The AIS system puts a data packet on a special VHF Marine Band channel that gives the ship's location, name, cargo, destination, origin, and other data. Harbor and shoreline radars track ship radar returns, and these returns are correlated in central databases across each coastline. These databases are cross-correlated so that we know exactly which ship is where, so long as it is in either radar or VHF radio range. Radar data can be supplemented by asynchronous data from other sensors, such as other radars, EO/IR sensors, or even visual sightings.
Navigation: Synopsis, Estimation Theory, Radar Trackers, Data Fusion, Top