Polarization Diversity Using Mutual Information

An information-theoretic criterion is introduced for waveform polarization type selection in a synthetic aperture radar (SAR) configuration. The criterion is based on the concept of mutual information (MI). Specifically, the MI criterion (MIC) minimizes the MI between the radar return signal at two distinct instants of time: one is the just-received return due to the last-transmitted pulse with known polarization type, and the other is the to-be-received return due to the pulse to be transmitted next with to-be-determined polarization type. In that manner, the polarization type selected for the pulse transmitted next will result in the collection of the largest amount of new information, as measured by a formal criterion. The MIC involves the radar system model and the probabilistic definition of the clutter, interference, and noise processes. In particular, when all these processes are proper Gaussian processes, the MIC attains a simple analytic form. The formulation and simulation-based results are presented in the context of a first-order radar system model, for simplicity, but the MIC can be extended to cover more complex models in a straightforward manner. In addition, the MIC is applicable to other radar modes and sensor types. The results presented show that the MIC is an effective method for polarization type selection in a SAR.