1. Introduction

The challenge of this chapter is to develop precise algorithms for modulation format recognition (MFR) with the highest identification accuracies in the presence of optical channel impairments. For that, we propose and demonstrate the use of ANN-based pattern recognition technique, trained by different feature-based approaches.

In the first method, we implement a new approach using ANN in conjunction with linear optical sampling (LOS) for direct and coherent systems at high data rates. Here, and in accordance to the IEEE 802.3 standards, we have considered the classification of 10 Gbps Non-Return-to-Zero On-Off Keying (NRZ OOK), 40 Gbps NRZ-Differential Quadrature Phase Shift Keying (DQPSK), 100 Gbps NRZ Dual-Polarized (DP)-QPSK, 160 Gbps DP-16 Quadrature Amplitude Modulation (16QAM) and 1 Tbps WDM-Nyquist NRZ-DP-QPSK digital modulation formats for high-speed communication systems. Numerical simulations demonstrate high identification precision in the

Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and eproduction in any medium, provided the original work is properly cited.

presence of different impairments, such as chromatic dispersion (CD), differential group delay (DGD) and amplified spontaneous emission (ASE) noise.

In the second method, we propose a novel technique of MFR algorithms using the time-frequency analysis, which is wavelet transform. In conjunction with ANN pattern recognition algorithm, this method is efficient for features extraction when it approximate both the signal envelop and frequency content. Continuous wavelet transform (CWT) is used to extract the classification features of 40 Gbps NRZ-OOK, and used three multi-carriers modulation formats namely 160 Gbps OFDM DP-16QAM, 400 Gbps Dual-Carrier (DC)-Polarization Division Multiplexed (PDM)-QPSK and 1 Tbps WDM-Nyquist NRZ-DP-QPSK. Through simulations, the proposed technique is able to classify these modulation schemes under different transmission impairments with high accuracy.
