Modulation Formats

Back to
Applications
Overview

Robust 100Gb/s Transmission using Coherent Pol-Mux QPSK

Description

Polarization multiplexing of QPSK (Pol-Mux QPSK) is a good candidate for the implementation of next generation high-speed transmission systems. It helps to reduce the requirements on electrical and opto-electrical components because it requires a symbol-rate of only one fourth of the bit-rate. Digital signal processing (DSP) combined with coherent detection has the potential to mitigate the impact of transmission impairments. This example illustrates the usage of DSP to mitigate the impact of phase noise*, chromatic dispersion (CD) and polarization-induced distortions (PMD and polarization cross-talk**).
* Phase and frequency mismatch between laser transmitter and local oscillator.
** Cross-talk between orthogonally polarized channels resulting from the misalignment between the states of polarization of the local oscillator and the detected signal.

Typical Results

The simulation setup is displayed in Figure 1. It consists of three 112 Gb/s Pol-Mux QPSK transmitters at 100GHz channel spacing, a transmission line and polarization-diversity receivers that include two 90° hybrids and a DSP unit (see Figure 2). The transmission line is composed of a SSMF, a polarization tracker (to control the alignment between the signal SOP and the polarization beam splitter at the receiver), a PMD emulator and a ASE source (to set the OSNR). After detection the signal is sampled at 56Gsamples/s (2 samples per symbol) and is passed to the DSP unit. Residual CD is compensated using a FIR filter with Nt taps. Polarization effects (PMD and Pol-Xtalk) are mitigated using a MIMO structure consisting of 5-tap FIR filters. The coefficients of the MIMO structure are optimized using the constant-modulus algorithm (CMA). The phase noise is corrected using a multi-symbol phase estimation (MSPE) technique based on the Viterbi & Viterbi algorithm. The estimated phase error is averaged over Ns symbols. The impact of CD for different number of taps Nt is displayed in Figure 3. The impact of polarization effects and phase noise (for different values of Ns) on the BER and constellation diagram is illustrated in Figure 4 and Figure 5, respectively.

Keywords

Coherent, Polarization Multiplexing (Pol-Mux), Quadrature Phase-Shift Keying (QPSK), Digital Signal Processing (DSP), Finite Impluse Response (FIR), Constant Modulus Algorithm (CMA), Multi-Symbol Phase Estimation (MSPE)