The research group Data Fusion for Array Sensors (RG DFA) is concerned with all kinds of interrelations between array signal processing and data fusion algorithms. This research area is a consequence of the special importance of arrays in modern sensor technology and the high potential of this kind of processing. Accordingly, this research field is overlapping in some fields with other research groups where array sensors play a role (RG Wide Area Surveillance and RG Localization and Navigation). Current research projects are:

Passive Coherent Localization (PCL) using GSM Transmitters

Passive Coherent Localization (PCL) of echoes of moving targets is achieved using mobile phone base stations as illuminators. This activity comprises the development of various experimental systems based on array antennas, development of signal processing algorithms for PCL with GSM and 3G emitters and development of tracking algorithms for all kinds of PCL systems (GSM, DAB, DVB-T, FM, in cooperation with Fraunhofer FHR)

Emitter Localization

One problem is ground based location of emitters with a large array antenna. Sophisticated high resolution space-time array processing methods are used to resolve multipath, to estimate the multipath delays and fitting these measurements with a propagation model from ray tracing using a digital terrain map. The model and data fitting is realized by Bayesian probability density matching. For location of emitters on ground from airborne sensors with small antenna array topics of calibration, error compensation algorithms, subspace-based methods and source number determination are considered (in cooperation with RG L&N). These results are also applied to passive localization of acoustic sources for underwater applications (in cooperation with NATO Underwater Research Center).

ABF Tracking

Modifications of tracking algorithms with adaptive arrays are considered that use special information available from adaptive monopulse signal processing (ABF: Adaptive Beam Forming).

GMTI Tracking

Tracking algorithms accounting for the special features of air-to-ground radars with clutter suppression by space-time adaptive processing (STAP) are developed. These algorithms exploit knowledge of the STAP clutter notch width as prior information for the tracker.

Radar Resource Management

For multifunctional radar, information is available at the tracking and information fusion level, which can be used to control and optimize the radar modes of operation (dwell time, revisit time, waveform design). This is a multi-objective multi-constraint optimization problem.