Signal Shaping for Optical Communications


Project Description

Optical fibres underpin our global information society and have experienced an astonishing evolution over the past four decades. Currently deployed commercial systems based on single-mode fibres can transmit data rates in excess of 10 Tb/s per fibre. Widely deployed for the global communications infrastructure, single-mode fibres currently carry more than 99% of the global Internet traffic and are a key component in the backbone networks for mobile telephony and the Internet. The continuation of this dramatic throughput growth has become constrained due to a power dependent nonlinear distortion in single mode fibres arising from a phenomenon known as the Kerr effect. The mitigation of this fibre nonlinearity is currently an area of intense research.

Signal (constellation) shaping can be used to increase the transmission rates of optical system. Signal shaping can be broadly classified into geometric and probabilistic shaping. In geometric shaping, the shape of the constellation is changed (e.g., by chosing constellation points from a nonequally spaced grid). In probabilistic shaping, the probabilities of the constellation points are chosen from a nonuniform distribution. While both techniques are well known and well understood for the linear AWGN channel, their design and potential are unclear for the nonlinear optical fibre channel. For the AWGN channel, the ultimate shaping gain is 1.53 dB. The ultimate shaping gain for the nonlinear channel is unknown, but some preliminary results show that in certain regimes this number could be as high as 1.88 dB.

In this project, we will investigate and develop techniques to increase the rates of optical communications by means of signal shaping. We will focus on designs based on an information-theoretic approach tailored to the nonlinear channel. In particular, we will use mutual information and generalised mutual information to characterise and design the optical transmission system. The obtained results will be of a fundamental nature, however, particular emphasis will be put on complexity reduction to enable real-time implementation of these algorithms.

Results and Publications

T. Fehenberger, A. Alvarado, G. Bocherer and N. Hanik, “On Probabilistic Shaping of Quadrature Amplitude Modulation for the Nonlinear Fiber Channel”, in Journal of Lightwave Technology, vol. 34, no. 21, pp. 5063-5073, Nov. 2016.

E. Agrell and A. Alvarado, “Signal Shaping for BICM at Low SNR”, in IEEE Transactions on Information Theory, vol. 59, no. 4, pp. 2396-2410, Dec. 2012.

T. Fehenberger, D. Lavery, R. Maher, A. Alvarado, P. Bayvel and N. Hanik, “Sensitivity Gains by Mismatched Probabilistic Shaping for Optical Communication Systems”, in IEEE Photonics Technology Letters, vol. 28, no. 7, pp. 786-789, Jan. 2016.


TU Eindhoven (TUE)