Source localization using airborne sensor arrays in the presence of manifold perturbations

This thesis studies localizing sources of electromagnetic energy using a sensor array whose manifold is only nominally known. The problem of source localization is studied in the context of an airborne array, making possible the observation of a ground source from multiple look angles.; External and self calibration algorithms are developed as a means of obtaining accurate source localization estimates when the sensor manifold is perturbed. External calibration assumes that the expected difference between the actual and modeled array manifold has already been established. Self calibration assumes that this expected difference is known only approximately, and relies on signals of opportunity in the environment to provide updates.; Several novel calibration algorithms are proposed, and their performance is tested on both conventional arrays and vector sensors. Conventional arrays are arrays which consist of identical sensors, whereas vector sensors are polarization-diverse arrays that are composed of different sensor types. The results indicate that significant performance gains are achieved with the use of these proposed calibration algorithms.