An Improved End-to-End Autoencoder Based on Reinforcement Learning by Using Decision Tree for Optical Transceivers.

In this paper, an improved end-to-end autoencoder based on reinforcement learning by using Decision Tree for optical transceivers is proposed and experimentally demonstrated. Transmitters and receivers are considered as an asymmetrical autoencoder combining a deep neural network and the Adaboost algorithm. Experimental results show that 48 Gb/s with 7% hard-decision forward error correction (HD-FEC) threshold under 65 km standard single mode fiber (SSMF) is achieved with proposed scheme. Moreover, we further experimentally study the Tree depth and the number of Decision Tree, which are the two main factors affecting the bit error rate performance. Experimental research afterwards showed that the effect from the number of Decision Tree as 30 on bit error rate (BER) flattens out under 48 Gb/s for the fiber range from 25 km and 75 km SSMF, and the influence of Tree depth on BER appears to be a gentle point when Tree Depth is 5, which is defined as the optimal depth point for aforementioned fiber range. Compared to the autoencoder based on a Fully-Connected Neural Network, our algorithm uses addition operations instead of multiplication operations, which can reduce computational complexity from 108 to 107 in multiplication and 106 to 108 in addition on the training phase.

Interconnecting data based on vortex beams by adjusting the ellipticity of a ring-core fiber.

In order to exchange data in a space-division multiplexing (SDM) system, a novel vortex-beam-based data interconnection concept, which is achieved by adjusting the ellipticity of a ring-core fiber, is proposed. A new ring-core fiber is also designed and fabricated for exchanging and propagating the data carried by first- or second-order vortex (orbital angular momentum) beams. The proposed scheme is not only analyzed and simulated in principle, but is also verified through experiments. The numerical results demonstrate that the vortex beams can be exchanged by appropriately adjusting the phase difference (with respect to the ellipticity of a ring-core fiber) between the even and odd vector modes. A new experimental platform is designed and established for the sake of investigating the feasibility of the proposed scheme. The experimental results are consistent with the results of the simulation, and demonstrate that the data carried by the first- or second-order vortex beams can be successfully switched with acceptable bit error rates (BERs) between the first-order vortex beams (L=1 or -1) or between the second-order vortex beams (L=2 or -2, left or right circular polarization), respectively. The measured BERs and constellation diagrams of 16-QAM are employed to evaluate the data exchange performance with respect to different cases (i.e., data exchange once or twice, and data exchange with or without crosstalk). The measured BERs and constellation diagrams also demonstrate that the performance degrades with increase in topological charge or crosstalk. The proposed scheme is flexible, simple, and reliable for data exchange in a SDM system.