Fractal interpolation of radar signatures for detecting stationary targets in ground clutter

Fractal geometry techniques are applied to the radar stationary target detection problem. Target and clutter signatures are fractalized via fractal interpolation, and their resulting fractal dimensions are determined. This fractal dimension was used to distinguish targets from clutter using a simple thresholding technique. This fractal dimension feature was used both independently and in combination with an even-channel, two-parameter CFAR (constant false alarm rate). The simulation´s best results were achieved with the combination algorithm set for an overall probability of false alarm of 10^-2, yielding a probability of detection of 0.75 for one of the targets considered. The results indicate that the fractal dimension (FD) feature used seems to be independent of amplitude. Thus, the FD information combined with traditional amplitude processing techniques will improve probabilities of detection