Blind 8PSK signals detection using discrete complex-valued Hopfield network

Abstract

The conventional neural networks which are limited to two-state neurons are not able to solve the problem of blind multi-state signal detection. A new algorithm based on discrete complex-valued Hopfield neural network(DCHNN) is proposed to detect MPSK signals blindly. Based on the energy minimization performed by the network, this paper suggests an direct design procedure that gives a Hermitian weight matrix such that each constellation signal in the MPSK state space is an attractive fixed point of the network‥ Simulation results show that the algorithm reaches the real equilibrium points and show satisfactory performance in detecting MPSK signals blindly.