We experimentally demonstrate the use of deep neural networks (DNNs) in combination with signals’ amplitude histograms (AHs) for simultaneous optical signal-to-noise ratio (OSNR) monitoring and modulation format identification (MFI) in digital coherent receivers. The proposed technique automatically extracts OSNR and modulation format dependent features of AHs, obtained after constant modulus algorithm (CMA) equalization, and exploits them for the joint estimation of these parameters. Experimental results for 112 Gbps polarization-multiplexed (PM) quadrature phase-shift keying (QPSK), 112 Gbps PM 16 quadrature amplitude modulation (16-QAM), and 240 Gbps PM 64-QAM signals demonstrate OSNR monitoring with mean estimation errors of 1.2 dB, 0.4 dB, and 1 dB, respectively. Similarly, the results for MFI show 100% identification accuracy for all three modulation formats. The proposed technique applies deep machine learning algorithms inside standard digital coherent receiver and does not require any additional hardware. Therefore, it is attractive for cost-effective multi-parameter estimation in next-generation elastic optical networks (EONs).
We propose a novel technique for modulation format identification (MFI) in digital coherent receivers by applying deep neural network (DNN) based pattern recognition on signals' amplitude histograms obtained after constant modulus algorithm (CMA) equalization. Experimental results for three commonly-used modulation formats demonstrate MFI with an accuracy of 100% over a wide optical signal-to-noise ratio (OSNR) range. The effects of fiber nonlinearity on the performance of MFI technique are also investigated. The proposed technique is non-data-aided (NDA) and avoids any additional hardware on top of standard digital coherent receiver. Therefore, it is ideal for simple and cost-effective MFI in future heterogeneous optical networks.
11 April 2014 in Journal of Optical Communications and Networking
Optics InfoBase is the Optical Society's online library for flagship journals, partnered and copublished journals, and recent proceedings from OSA conferences.
We experimentally demonstrate simultaneous optical signal-to-noise ratio (OSNR) monitoring and modulation format identification (MFI) in heterogeneous fiber-optic networks by using principal component analysis (PCA) and statistical distance measurement based pattern recognition on scatter plots obtained through asynchronous single channel sampling (ASCS). The proposed technique enables OSNR monitoring for several commonly-used modulation formats with mean OSNR estimation error of 1 dB and without requiring any information about the signal type during the online monitoring process. In addition, it successfully demonstrates the identification of unknown modulation formats of the received signals with an overall accuracy of 98.46%. The effects of chromatic dispersion (CD) on the performance of proposed technique are also analyzed. Due to the use of a single low-speed asynchronous sampling device in the proposed technique, the implementation complexity and cost of the monitoring devices can be significantly reduced.