Neural Network Equalizers and Successive Interference Cancellation for Bandlimited Channels with a Nonlinearity

Neural networks (NNs) inspired by the forward-backward algorithm (FBA) are used as equalizers for bandlimited channels with a memoryless nonlinearity. The NN-equalizers are combined with successive interference cancellation (SIC) to approach the information rates of joint detection and decoding (JDD) with considerably less complexity than JDD and other existing equalizers. Simulations for short-haul optical fiber links with square-law detection illustrate the gains of NNs as compared to the complexity-limited FBA and Gibbs sampling.

Phase Matching for Multimode Four-Wave Mixing in Few-Mode Fibers and Nano-Rib Waveguides

We compare phase matching for four-wave mixing using one, two, three, and four waveguide modes. For the comparison, we use numerical optimizations and an estimate of the generated idler power. We present results for few-mode fibers and nano-rib waveguides and show that for both waveguide types, four-wave mixing bandwidths and idler powers are best for one- and two-mode operation and that four-mode four-wave mixing is not feasible at all. Some nano-rib waveguides support three-mode four-wave mixing, albeit with much reduced bandwidth and reduced idler power.

Comparison of PAM-6 Modulations for Short-Reach Fiber-Optic Links with Intensity Modulation and Direct Detection

PAM-6 transmission is considered for short-reach fiber-optic links with intensity modulation and direct detection. Experiments show that probabilistically-shaped PAM-6 and a framed-cross QAM-32 constellation outperform conventional cross QAM-32 under a peak power constraint.

Experiments on Bipolar Transmission with Direct Detection

Achievable information rates of bipolar 4- and 8-ary constellations are experimentally compared to those of intensity modulation (IM) when using an oversampled direct detection receiver. The bipolar constellations gain up to 1.8 dB over their IM counterparts.