When using multichannel synthetic aperture radar (MSAR) to perform moving target detection, the coherence between the received signals is closely linked to the detection capacities of the system. In airborne MSAR context, because of the platform movement, the antenna patterns, as well as the phase shifts between the channels, evolve in different ways during the integration time. In this paper, these dissimilarities are shown to cause a degradation of coherence, and hence of detection capacities. A method to enhance the coherence is then developed. This method is based on the estimation of the steering vector. It is applied on two sets of real data, and the obtained results validate the method.

Detecting and locating moving targets are key issues for radar ground surveillance. Because of its attractive resolution, the use of SAR imaging for this purpose is of prime interest. However, SAR was designed to image non-moving targets, and thus cannot be used as it stands. Some research has been carried to develop simultaneous SAR/STAP processing, because of STAP capacity to detect moving targets. In this paper, a new algorithm for detecting moving targets and for estimating their true location and their velocity is presented. Results obtained on real data are also presented.

An original application of the weighted deflection is proposed for radar target classification by quadratic filters. An explicit formulation of optimal filters is derived. We analyse the impact of the weighting parameter on real data recognition, and show that performances are better when the deflection coincides with the Fisher ratio.