A blind equalization algorithm based on bilinear recurrent neural network

In this paper, we propose a new blind equalization algorithm based on bilinear recurrent neural network. In the first part, a new transmission function and cost function is designed for the neural network. In the second part, the new neural network is used to blind equalization algorithm. Results of the simulation show that our algorithm can obtains better convergence performance and lower bit error rate (BER) than traditional constant modulus algorithm (CMA).