An Adaptive step size Stokes Space method for polarization demultiplexing

This paper studies the polarization demultiplexing technology in the coherent detection of 100G DP-QPSK optical signals. An Adaptive step size Stokes space method for polarization demultiplexing is proposed, which solves the problem of insufficient convergence caused by the constant step size of the Iterable Stokes space method. The original Iterable Stokes space method has no singularity problem, fast convergence speed and better polarization state tracking performance, but its iterative step length is fixed. When setting the step length, it is necessary to find a balance between the convergence speed and the convergence stability. In this paper, the Iterable Stokes space method is improved. The new algorithm proposed retains the advantages of the Iterable Stokes space method and realizes the adaptive update of the step size. The simulation results show that compared with the Iterable Stokes space method, the proposed new algorithm adds a small amount of operations, but the convergence speed is doubled, the convergence is more stable, and the polarization state tracking performance is also improved.