Space-based AIS multi-user separation based on the improved complex value FastICA

The coverage of the space-based automatic identification system (AIS) is wide and it can be used to monitor the global ships effectively. It also brings a new challenge that the collision of AIS signals from different cells results in loss of all collided message. In order to solve this problem, this paper proposes a multi-user separation scheme based on improved complex value FastICA. In this paper, we give full consideration to the communication mode and signal characteristics of the system. Meanwhile, a step factor is introduced to the complex value FastICA to reduce the algorithm's sensitivity to initial value. In order to reduce the amount of calculation, the introduced step factor searched based on Wolfe-Powell. Simulation results show that the proposed scheme can separate the collided signals effectively.