Seb Savory

ABSTRACT This paper serves to highlight the gains in SNR margin and/or data capacity that can be achieved through a proper optimization of the transceiver parameters, for example, launch power, modulation format, and channel allocation. A simple quality of transmission estimator is described that allows a rapid estimation of the signal quality based on ASE noise and nonlinear interference utilizing the Gaussian noise model. The quality of transmission estimator was used to optimize the SNR and maximise the data throughput of transmission signals in a point-to-point link by adjusting the launch power and modulation format. In a three-node network, the launch power and channel allocation were adjusted to minimise the overall effect of nonlinear interference. This paper goes on to show that by optimizing the transceiver modulation format as part of the channel allocation and routing problem gains in network data throughput can be achieved for the 14-node NSF mesh network.

We report experimental results of 8×40 Gbit/s polarization-channel-interleaved carrier-suppressed RZ transmission over 2×160 km NDSF, with 0.8 bit/s/Hz spectral efficiency. Average system linear Q of 6.5 (231-1 PRBS) and tolerance of at least 50% degradation in orthogonality were demonstrated.

The deployment of coherent transceivers in legacy networks requires significant investment in installation. We propose a method enabling autonomous (re-)configuration of an optical channel, which would be advantageous in legacy networks and necessary in proposed future networks utilizing a flexible frequency grid and software defined components such as reconfigurable optical add drop multiplexers (ROADM). We consider potential interfering optical channels propagating with the prospective channel along part of the fiber link which are dropped before arrival at the receiver. The method uses a commercially available line card transmitting a 40Gbit/s polarization multiplexed quadrature phase shift keying (PM-QPSK) probe channel to characterize an optical channel. The power of the nearest neighboring channels is then inferred by examining its bit error rate (BER) which is impacted by cross phase modulation (XPM) from the aggressor channels. In a 4 node network with 2 ROADMs using up to 6...

ABSTRACT Fast wavelength switching OFDM transceiver enables the coherent reception of 2-burst channels within a 1dB penalty after 800km transmission. Burst detection and variable path-history compensation are performed using inherent OFDM synchronization symbols and cyclic prefix.

... case of downstream transmission (receiver at the ONU), we find that, compared with the single channel configuration, nonlinear penalty is incurred when the transmission power is above 8 dBm per ... 3(a) shows a sensitivity benefit due to preamplifying the receiver of 7.1 dB. ...

ABSTRACT Coherent receivers offer a potential solution for implementing a high-capacity, long-reach (up to 100 km) passive optical network (LR-PON), due mainly to their high sensitivity, frequency selectivity, and bandwidth efficiency. When using coherent receivers, received signals can be post-processed digitally to mitigate the specific impairments found in access networks and, additionally, relax the optical complexity requirements of the coherent receiver. However, the digital signal processing must itself be low complexity in order to minimize the overall complexity and power consumption of the optical network unit (ONU). This paper focuses on the impact of reduced complexity equalization algorithms on receiver sensitivity in a LR-PON. It is found that a cascade of linear filters can be combined into a single, truncated, linear, adaptive filter with negligible impact on receiver sensitivity. Additionally, by utilizing a multiplier-free tap weight update algorithm, the overall complexity of a digital coherent receiver can be significantly reduced, making it attractive for use in an ONU. Matched filtering, chromatic dispersion compensation, and polarization tracking are all performed by the adaptive equalizer. The performance of this low-complexity, multiplier-free equalizer is experimentally verified for 3 GBd polarization division multiplexed quadrature phase shift keying (12 Gbit/s) in both a back-to-back configuration and transmission over 100 km standard single-mode fiber.

ABSTRACT This paper presents a hardware-efficient carrier phase estimator with high-linewidth tolerance for 16-QAM optical coherent systems. The laser phase noise is estimated using quaternary phase-shift keying (QPSK) partitioning complemented with a low complexity angle-based barycenter approximation as opposed to the classical Viterbi and Viterbi algorithm. The various stages necessary for partitioning and removing the modulation on the received symbols for carrier phase recovery are presented. We show that the phase offset in the middle ring for a 16-QAM constellation can be removed through a simple comparison with the symbols lying on the inner and outer rings of the constellation thus enabling all the symbols to be efficiently utilized for carrier phase recovery. We assess the performance of different filter structures for 16-QAM with filter half width 8 and 16. Simulation results demonstrate that combined linewidth symbol duration product $\Delta \nu \cdot T_s$ of $10^{-4}$ is tolerable at the target BER of $10^{-2}$ and $10^{-3}$ when using the barycenter algorithm. Finally, carrier phase recovery in a 16-QAM experiment is investigated to validate the performance of the proposed algorithm.

ABSTRACT We investigate the use of a fast switching DSDBR tunable laser in a coherent receiver to enable microsecond sweeping time over the C-band. Thomson's Multitaper method is used to estimate spectral slices which are then digitally stitched to form the complete scanned spectrum.

ABSTRACT A commercially available digital supermode distributed Bragg reflector tunable laser is employed as a fast wavelength switching local oscillator (LO) in a dual polarization (DP) 16-quadrature amplitude modulation (16QAM) coherent burst mode receiver. A digital coherence enhancement technique is used to compensate both the Lorentzian and non-Lorentzian distributed phase noise of the tunable LO laser. It is shown that differential decoding is not sufficient to overcome the substantial bit errors caused by the LO laser phase noise. However, the coherence enhancement technique enables the reception of low symbol rate DP-16QAM bursts, with an average optical signal to noise ratio penalty of 3.5 dB observed relative to theory at the forward error correction threshold $(1.5times 10^{-2})$.

ABSTRACT This letter presents the carrier phase recovery for 16-ary quadrature amplitude modulation (16-QAM) optical coherent systems using the quaternary phase-shift keying (QPSK) partitioning with sliding window averaging and differential decoding. We assess the increase in linewidth tolerance achievable with sliding window averaging as opposed to block averaging. Simulation results demonstrate that combined linewidth symbol duration product, $Deltanucdot T_{s}$ , of $10^{-4}$ is tolerable at the target bit error ratio (BER) of $10^{-2}$ and $10^{-3}$ for a penalty of 0.6 and 0.8 dB, respectively, compared with the theoretical limit with differential decoding. The impact of analog-to-digital converter (ADC) resolution on the performance of the QPSK partitioning algorithm is also investigated. Finally, the performance of the algorithm using the measured phase noise for a distributed feedback (DFB) laser is presented for different values of $Deltanucdot T_{s}$ . We show that for $Deltanucdot T_{s}>10^{-4}$ , the penalty of block averaging is ${>}{rm 0.5}~{rm dB}$ with respect to sliding window averaging at the target BER of $10^{-3}$ with the measured phase noise. The degradation associated with block averaging at the target BER of $10^{-2}$ is shown to be less significant compared to sliding window averaging.

ABSTRACT In-phase and quadrature skew inside a coherent receiver, caused by misalignments, can severely limit performance. Future 400G systems employing higher order modulation formats and Nyquist filtering are especially sensitive to this. A blind adaptive equalizer is proposed to track and compensate the skew while also performing matched filtering. Performance is investigated in simulation for 56 GBd polarization division multiplexed 16-quadrature amplitude modulation (PDM-16QAM) and experimentally for 6 GBd PDM quadrature phase shift keying (PDM-QPSK) and PDM-16QAM, where compensation for delays up to half a symbol period are demonstrated without penalty. At 30% delay skew, the maximum skew recommended by Optical Internetwork Forum (OIF) for PDM-QPSK, a gain of 1.6 dB for PDM-QPSK, and a gain of more than 5 dB for PDM-16QAM is observed compared with using a conventional equalizer.

ABSTRACT Nyquist-WDM PDM-QPSK transmission over standard SMF-28 fibre with novel distributed URFL amplification is reported for the first time. Transmission over 6157 km in a recirculating loop, and 320 km in an unrepeatered link without a ROPA were demonstrated.

This letter explores the Gram-Schmidt orthogonalization procedure (GSOP) for compensation of quadrature imbalance in an optical 90deg hybrid. We present computer simulations for an optical QPSK communication system using a digital coherent receiver and investigate the impact of quadrature imbalance on the required optical signal-to-noise ratio for the receiver and the frequency estimation algorithm. We then demonstrate the improvement which

The feasibility of a colorless coherent PON operating at 10 Gbit/s is verified using monolithically integrated C-band tunable lasers as sources for the signal and local oscillator. Using parallel digital postprocessing, a sensitivity below −36 dBm is demonstrated.