Effective array calibration criterions for radar low-angle tracking

Low-angle radar tracking is, and has long been, one of the most important problems in the field of radar. Due to the fact that target echoes that are received directly and via the sea surface are unresolved or merged in time and frequency, the performance of classical monopulse tracking radar is seriously degraded. A considerable amount of research has been done on array signal processing techniques to separate a target from its image and to provide an accurate estimate of the elevation of the target. To date, most array calibration techniques do not work perfectly. With a view to practical use, an effective calibration method should have a real-time capability and a ´blind´ capability, which means that no a priori knowledge of the signal directions and array errors is needed and the combined effect of all possible kinds of error sources can be tackled together. Two simple, yet powerful, calibration criteria are presented to attempt to fulfil this requirement. The idea is to use as much a priori information in low angle tracking as possible to find some mathematical characteristics of the ideal array data, which are then used as calibration criteria to achieve optimal fitting between the received and ideal data sets.