Research On Time-domain Hybrid Optical Signal Polarization Demultiplexing And Modulation Format Identification

The evolution of optical communication networks goes towards the maximum time-flexibility. In order to deal with the fast growth of the Internet traffic and the emerging data center architecture, next-generation transceiver is required to support adaptive spectral efficiency(SE). Time domain hybrid modulation format is one of promising technologies to achieve continuous tuning on spectral efficiency or bit rate. Using the polarization property of optics transmitted in a single mode fiber, Polarizaton multiplexing technique aims at creating two independent communication channels over the transmission wavelength by using the independent and mutually orthogonal polarization, which can double the capacity of optical fiber communication system and increase spectrum efficiency. The two polarizations can be affected by birefringence, which will cause crosstalk between each other. The most important problem to be solved in polarization multiplexing system is eliminating the crosstalk between the received signals.With the advent of reconfigurable transmitters capable of signal generation using arbitrary coherent optical modulation format, it is no longer possible to ensure that the receiver unit will know the incoming modulation format in advance. Modulation format recognition(MFR) is essential to guarantee that signals, which are optimally acquired and demodulated. An advantage of Stokes space is that polarization mixing, carrier frequency offset and phase offset do not affect the 3-dimensional(3D) representation of the signal in the Poincaré sphere. By utilizing the DSP capabilities of a digital coherent receiver,we use Stokes space representation of the signal and employ a machine learning algorithm known as variational Bayesian expectation maximization(VBEM) for Gaussian mixture models(GMM).The technique is based entirely on the observation of samples in Stokes space, does not involve demodulation and is modulation format independent. The data in Stokes space is used to find the best fit plane and the normal to it which contains the origin. We have studied the stoke space polarization demultiplexing performance based on high speed coherent optical detection of 100 Gb/s PDM-QPSK system and 224 Gb/s PDM-16 QAM system. The technique has been demonstrated for polarization demultiplexing of arbitrary complex-modulated signals. We have also simulated time-domain hybrid signal and discussed the performance under different power ratios and symbols situation. Finally, we discuss the convergence rate base on stokes space polarization demultiplexing.We study on modulation format recognition based on variational Bayesian expectation maximization algorithm. The method is based on Stokes space signal representation and clusters modulation format in the three-dimensional space. QPSK?8PSK ? 8QAM and 16 QAM modulation formats are successfully demonstrated numerically before polarization demultiplexing. We have also studied received signals after polarization demultiplexing and propose a simple modulation format identification base on VBEM. Simulation results demonstrate successful QPSK?16QAM and 32 QAM among different modulation formats signals in practical system OSNRs.