Low-complexity modulation format identification scheme via graph-theory in digital coherent optical receivers

Abstract

Efficient modulation format identification (MFI) scheme is necessary for the flexible digital coherent transceivers in the next-generation elastic optical networks. In this paper, an innovative graph-theory based MFI scheme is presented, which transfers the trajectory and coordinate information of adjacent received symbols to the graph domain. The proposed MFI scheme contains two crucial stages, i.e. discriminated-feature extracted stage and the minimal acute-angle comparison stage. In the discriminated-feature extracted stage, a uniform identify grid model is established in the first quadrant of the constellation diagram, and a concise graph structure is conducted according to the coordinate and trajectory information of the adjacent received symbols. The eigenvector associated with the largest eigenvalue is selected as the discriminated-feature of the corresponding modulation format signal. In the minimal acute-angle comparison stage, the modulation format type of unknown signal is identified by calculating the acute-angle between the discriminated-features of the canonical modulation formats signals and the unknown one. The effectiveness of the proposed MFI scheme is demonstrated via 28 GBaud numerical simulations with polarization-division multiplexing (PDM)-quadrature phase shift keying (QPSK), PDM-8quadrature amplitude modulation (QAM), PDM-16QAM, PDM-32QAM and PDM-64QAM signals. The results show that high identification accuracy can be achieved over wide optical signal-to-noise ratio (OSNR) ranges. Meanwhile, we also discuss the influence of the residual chromatic dispersion (CD), launching optical powering, different group delay (DGD) to the proposed MFI scheme. Finally, the proposed MFI scheme is further verified by 20 GBaud PDM-QPSK/8QAM/-16QAM/-32QAM long-haul transmission experiments. The proposed MFI scheme with low computational complexity would certainly provide a superior convenience to the future elastic optical networks.