Blind Separation of Chaotic Signal Based on Random Set Model

Traditional blind separation methods perform well when the number of source signals is constant, but in practice, the number is usually time-varying. In this case, traditonal methods can hardly achieve separation. In this paper, a method is presented to deal with the problem of blind separation for chaotic signals in a dynamic environment. This paper first establishes dynamic system model that based on random set theory(RST), then presents an algorithm based on particle filtering(PF) to separate the chaotic signal. Simulation results show that the proposed algorithm can track the number of source signals and has good separation performance.