Fusion center with neural network for target detection in background clutter

Analysis of radar signals for target detection in background clutter involves the use of different algorithms. These algorithms provide different levels of detection probability and false alarms as a function of the clutter present. This paper provides a solution to the problem of selecting the appropriate algorithm for target detection in background clutter with high probability of detection and low false alarms. The approach is based in parallel execution of CA-CFAR (cell averaging constant false alarm rate), GO-CFAR (greatest off) and SO-CFAR (smallest off) algorithms and a fusion center based on a neural network with different fusion rules. Results with simulated and real data are presented and discussed.