A new robust neural network method for coherent interference rejection in adaptive array systems

An approach for narrowband interference rejection is presented using a multilayer neural network with a three-element radar array system. The simulation compares the performances of the adaptive array systems for the conventional least-mean-square (LMS) network and the multilayer neural network. The results verify the feasibility of the multilayer neural network by means of the back-propagation (BP) algorithm for adaptive radar systems. These results show many advantages over the conventional LMS method for the following problems: coherent interference rejection, simultaneous multiple target detection, influence of the convergence speed by the input signal power, computational complexity, real-time processing and robustness