Optical performance monitoring (OPM) is the estimation and acquisition of different physical parameters of transmitted signals and various components of an optical network. OPM functionalities are indispensable in ensuring robust network operation and plays a key role in enabling flexibility and improve overall network efficiency. We review the development of various OPM techniques for direct-detection systems and digital coherent systems and discuss future OPM challenges in flexible and elastic optical networks.

Optical Performance Monitoring: A Review of Current and Future Technologies

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).

Joint OSNR monitoring and modulation format identification in digital coherent receivers using deep neural networks

We propose a transfer learning assisted deep neural network (DNN) method for optical-signal-to-noise ratio (OSNR) monitoring and realize fast remodel to response to various system parameters changing, e.g. optical launch power, residual chromatic dispersion (CD) and bit rate. By transferring the hyper-parameters of DNN at the initial stage, we can fast response to the channel variation with fewer training set size and calculations to save consumptions. For feature extraction processing, we use amplitude histograms of received 56-Gb/s QPSK signals as the input for DNN at the initial stage, which shows the root mean squared error (RMSE) of OSNR estimation is less than 0.1 dB with the OSNRs ranging from 5 to 35 dB. Then, we change several system parameters and find superior capabilities of fast remodeling and data resource saving with the proposed method. The required training epochs have about four times reduction, and the required training set size is only one-fifth compared to retraining the network without any accuracy penalty. The DNN assisted by transfer learning can save resources and will be beneficial for real-time application on OSNR estimation.

Transfer learning assisted deep neural network for OSNR estimation

The trade-off between more user bandwidth and quality of service requirements introduces unprecedented challenges to the next generation smart optical networks. In this regard, the use of optical performance monitoring (OPM) and modulation format identification (MFI) techniques becomes a common need to enable the development of next-generation autonomous optical networks, with ultra-low latency and self-adaptability. Recently, machine learning (ML)-based techniques have emerged as a vital solution to many challenging aspects of OPM and MFI in terms of reliability, quality, and implementation efficiency. This article surveys ML-based OPM and MFI techniques proposed in the literature. First, we address the key advantages of employing ML algorithms in optical networks. Then, we review the main optical impairments and modulation formats being monitored and classified, respectively, using ML algorithms. Additionally, we discuss the current status of optical networks in terms of MFI and OPM. This includes standards, monitoring parameters, and the available commercial products with their limitations. Second, we provide a comprehensive review of the available ML-based techniques for MFI, OPM, and joint MFI/OPM, describing their performance, advantages, and limitations. Third, we give an overview of the exiting ML-based OPM and MFI techniques for the emerging optical networks such as the new fiber-based networks that use future space division multiplexing techniques (e.g., few-mode fiber), the hybrid radio-over-fiber networks, and the free space optical networks. Finally, we discuss the open issues, potential future research directions, and recommendations for the potential implementation of ML-based OPM and MFI techniques. Some lessons learned are presented after each section throughout the paper to help the reader identifying the gaps, weaknesses, and strengths in this field.

/pdf/machine-learning-techniques-for-optical-performance-47wf3d2fwv.pdf

Machine Learning Techniques for Optical Performance Monitoring and Modulation Format Identification: A Survey

We experimentally demonstrate a modulation format identification (MFI) and optical signal-to-noise ratio (OSNR) monitoring method for digital coherent receivers by using the specific features of received signals' density distributions in Stokes axes combined with deep neural networks (DNNs). The features of received signals' density distribution fitting curves in S1 and S2 axes depend on the signal's modulation format and OSNR. The proposed technique extracts the features of these fitting curves' first-order derivation for MFI and OSNR monitoring, in order to improve the probability of format correct identification and OSNR estimation accuracy. Experimental results for 28Gbaud/s polarization-division multiplexing (PDM) quadrature phase-shift keying (QPSK), PDM 8 quadrature amplitude modulation (PDM-8QAM), PDM-16QAM, and 21.5Gbaud/s PDM-32QAM signals demonstrate OSNR monitoring over back-to-back transmission with mean estimation standard errors (SEs) of 0.21dB, 0.48dB, 0.35dB and 0.44dB, respectively. The MFI results over back-to-back transmission show that 100% identification accuracy of all these four modulation formats are achieved at the OSNR values lower or equal to their respective 7% forward error correction (FEC) thresholds. Similarly, 100% identification accuracy also is obtained for PDM-QPSK, PDM-8QAM, PDM-16QAM, and PDM-32QAM after 2000km, 2000km, 1040km, and 400km standard single-mode fiber (SMF) transmission within practical optical power ranges launched to the fiber, respectively.

Modulation format identification and OSNR monitoring using density distributions in Stokes axes for digital coherent receivers.

We propose a novel method of in-band estimation of optical signal-to-noise ratio (OSNR) using a digital coherent receiver, where OSNR is determined from second- and fourth-order statistical moments of equalized signals in any modulation format. Our proposed method is especially important in recently-developed Nyquist wavelength-division multiplexed (WDM) systems and/or reconfigurable optical-add/drop-multiplexed (ROADM) networks, because in these systems and networks, we cannot apply the conventional OSNR estimation method based on optical-spectrum measurements of the in-band signal and the out-of-band noise. Effectiveness of the proposed method is validated with computer simulations of Nyquist-WDM systems and ROADM networks using 25-Gbaud quadrature phase-shift keying (QPSK) and 16 quadrature-amplitude modulation (16-QAM) formats.

In-Band Estimation of Optical Signal-to-Noise Ratio From Equalized Signals in Digital Coherent Receivers

We propose and experimentally verify a fully-blind scheme for equalization and carrier-phase recovery in QPSK coherent optical receivers, which is based on the modified CMA. Its computational complexity is much less than the conventional approach.

Blind equalization and carrier-phase recovery in QPSK coherent optical receivers based on modified constant-modulus algorithm

Digital coherent optical receivers enable equalization of fiber transmission impairments and carrier-phase estimation in digital domain. In this paper, a novel equalization and carrier-phase recovery algorithm, both based on modified constant-modulus algorithm (CMA), is proposed for polarization-division multiplexed quadrature phase-shift keying (PDM-QPSK) systems. The proposed scheme is validated with 40-Gbits/s PDM-QPSK transmission experiments.

Modified CMA based blind equalization and carrier-phase recovery in PDM-QPSK coherent optical receivers

We propose and demonstrate least radial distance based carrier-phase recovering method for 16-QAM coherent optical systems. Through numerical analysis, we found proposed scheme has less than 2dB SNR penalty when laser linewidth-symbol-duration-product product Δv Ts ≅ 1 × 10-4 range.

Saifuddin Faruk

Papers

In-Band Estimation of Optical Signal-to-Noise Ratio From Equalized Signals in Digital Coherent Receivers

Blind equalization and carrier-phase recovery in QPSK coherent optical receivers based on modified constant-modulus algorithm

Modified CMA based blind equalization and carrier-phase recovery in PDM-QPSK coherent optical receivers

Least radial distance based carrier phase recovery for 16-QAM coherent optical systems