SAR PROCESSING: AT THE HEART OF THE SAR TECHNIQUE
In this chapter we will explore the basic and most significant features of radar imagery. In a first part, we will introduce the general principles of synthetic aperture radar (SAR), highlighting some specific effects strongly related to the physics and geometry involved. In a second part, we will develop the key role of the frequency representation with its manifold advantages: strong physical signification, access to more efficient computing schemes, specific resampling procedures. In this part we will also show that in some cases, working in the time domain remains preferable. In the third part we will review the SAR synthesis algorithms in order to give their specific flavors rather than going in full details: what is important is the understanding of the main architectural choice attached to each design, together with its advantages and drawbacks. The fourth part is devoted to the system constraints, which is the way to accommodate the many geometric and time constraints in order to optimize such or such features of the radar images. The fifth part will focus on the geometric properties of radar imagery and the equations which govern the positioning of radar pixels on the ground. Some methods for topography reconstruction are also briefly evoked in this part. Finally, we address the specific processing and co-registration problems raised by bistatic systems, and we propose a geometric modeling for bistatic observations.