Nonlinear Power Transfer Function Based Optical Modulation Format Identification

With the development of coherent optical communication technology, more and more advanced modulation formats are widely used and the data rates increase continuously. With the Internet service tends to diversification, it is envisaged that the future fiber-optic networks will be heterogeneous in terms of modulation formats and data rates in nature. Thus modulation format and data rate identification in heterogeneous fiber-optic networks is important for the network optimization, management and adaptive signal reception.In this thesis a simple and cost-effective all-optical technique based on four-wave mixing(FWM) in highly nonlinear optical fibers(HNLFs) for modulation format identification(MFI) is proposed and studied. First it is revealed that the signal pulse peak power variations against chromatic dispersion(CD) are dependent on the modulation formats of the signal. Second, based on the FWM theory, the theoretical model the MFI device is setup. Utilizing the nonlinear power transfer function(NPTF) provided by the FWM the peak power variation is mapped on the variations of the average power of output idler wave. Furthermore the NPTF is optimized according the characteristics of the widely used modulation formats to enhance the output contrast of the output results for effective feature recognition. Based on these results, the feature curves for the ten widely used modulation formats are obtained. In addition, he polarization-multiplexed signals are correctly identified. Using very simple cross correlation algorithm or the support vector machine(SVM) algorithm the ten modulation formats can be identified correctly. Finally the identification method of the data rates is also discussed. The MFI method proposed doesn't require any prior knowledge of the signal and has no limit on the data rates. It is applicable to on-off keying and advanced modulation format signals with CD and OSNR degradations. The method is also simple and cost effective because no high speed device or complicate data processing is needed.