RECOGNIZING CHAOS IN RADAR IMAGES

Many physical processes can be described in terms of nonlinear dynamic s, i.e. chaos. It is thus of interest to the experimentalist to discov er whether an observed time series can be modelled as a chaotic proces s. Techniques for recognizing chaos exist which rely on measuring the fractal dimension of the underlying strange attractor but may be disru pted by the presence of noise on the experimental data. This paper fir stly examines claims that the fluctuating radar cross section of the s ea surface can be recognized as a low-dimensional chaotic process and then goes on to consider whether identification of such a process is p ossible in the presence of additive or multiplicative noise. The relev ance of these results is discussed with regard to the problem of recog nizing chaos in high-resolution radar images which are corrupted by mu ltiplicative speckle noise.